By Kieran Ryan, 2024
OpenAI is led by Sam Altman, one of the most powerful people in the world of technology. Sam is the former President of Y Combinator and co-signed (staking his reputation on) by Paul Graham himself. Paul included Sam alongside Steve Jobs (Founder of Apple) and Paul Buchheit (Founder of Gmail). It’s wild to think that Paul published that essay in April 2009 since Sam has accomplished so much since then. Sam’s network has helped him recruit A+ co-founders and executives, including Greg Brockman, Mira Murati, and Brad Lightcap. Sam’s YC network will give OpenAI a distribution edge to make inroads with top startups. You can see this developing in real time as OpenAI announces new partnerships with top YC startups, such as these announcements with Brex, Instacart, and Stripe.
OpenAI beat Google and others who have been building artificial intelligence products internally for years to market. The first-mover advantage helped them achieve record-breaking traction metrics. OpenAI’s ChatGPT surpassed 100 million users two months after launch, making it the fastest-growing consumer app ever. Additionally, the first-mover advantage gives them a huge marketing edge. It was the first product to delight users, even if Google Bard and alternatives grade similarly, and ChatGPT is now synonymous with any mention of artificial intelligence.
Microsoft gives OpenAI the cloud computing infrastructure to power its research, products, and API services. The amount of compute needed to train AI models has been increasing exponentially, so without a partner like Microsoft, it’s unlikely a startup, even as well-funded as OpenAI, could compete with Meta or Google. The partnership with Microsoft also gives OpenAI more distribution since Microsoft is using OpenAI’s technology to build out its Azure infrastructure, power Bing, and supercharge other products.
Big companies can only exist in big markets. I’ve seen projections for the AI market to be anywhere from 2 trillion to 15 trillion by 2030. There is no clear winner yet as consumers have shown a willingness to experiment with different products to satisfy their needs. Still, the category winner will likely be worth 150 billion+ given the projected market size, surpassing other category winners like Uber (95.59 billion market cap), Airbnb (95.13 billion market cap), and DoorDash (32.83 billion market cap).
In the last year or so, Stripe has launched Financial Connections, a product that directly competes with Plaid, and a charge card program, a product that directly competes with Klarna and Ramp. With access to so many existing customers and the 1,000 new businesses that sign up on Stripe each day, Stripe will continue to use its product distribution to launch and scale new products, some that are copycats, to its customer base. Everyone copies everyone, and Stripe could do it the most successfully in Fintech.
Stripe’s 2022 annual update shared that they processed $817 billion in transactions, up 26% from $640 billion in 2021. Stripe is loved by large enterprise customers, with 100 customers handling more than $1 billion in payments with Stripe every year, and startups, more than 50% of these 100 customers have grown revenue by 10x, and 25% have grown revenue by 100x while on Stripe. Stripe is acquiring customers early with Stripe Atlas and helping them scale with tools that drive integer percentage point increases to top-line revenue in an industry that sees basis points standardly. While growth is still incredibly good for a company of Stripe’s size, it is slowing down a bit — the 26% increase in transaction volume is down from the 60% increase the year before. Its new product lines, like Financial Connections and its charge card program, should become exciting new revenue lines. Still, another thing Stripe is doing to reinvigorate its growth is signing large enterprise payment partnerships with companies like Amazon, Uber, Microsoft, and BMW.
Look, as good as Stripe is, it isn’t shielded from changes in the global economy. Like most startups, Stripe over-hired and, as a result, has had to conduct some layoffs, including 14% of its staff in November 2022 and a few dozen in June 2023. Additionally, Stripe’s valuation was cut down from $95 billion to $50 billion in March 2023. Even with a few bumps and bruises along the way, I was slightly surprised, since joining a startup at such a high valuation leaves minimal equity upside, to learn that Stripe is still a very desirable place to work. As of July 2023, Stripe ranks 3rd on my startup leaderboard for most in-demand places to work, including receiving votes from a Director of Engineering at a Series B startup backed by Y Combinator and a Partner at an early-stage venture capital firm.
In Q4 2022, The Setter 30, a report on the most in-demand secondary shares based on a survey of 500 late-stage investors, ranked Stripe second, only behind SpaceX—this marked Stripe’s 8th consecutive quarter of ranking in the top 2. Late-stage investors are still confident that Stripe can produce long-term growth and returns for shareholders and ultimately be the category winner.
Linear has created a cult-like following from its customers who tell their friends about the product and help them increase adoption within the company. Multiple friends, both product managers, and engineers, have proactively confessed their love for the product to me in conversation without even being prompted.
Linear is backed by Sequoia Capital, Index Ventures, and Dylan Field. They last raised in 2020 and have not needed to raise again because they are profitable and growing exponentially. Their investors will help them fundraise in the future if needed and give them an edge when hiring executive talent and for distribution. On Linear’s customer page, I noticed several startups who share the same investors, like Loom (backed by Sequoia Capital and Dylan Field), Retool (backed by Sequoia Capital), Watershed (backed by Sequoia Capital), and Alan (backed by Index Ventures).
In May 2023, Linear landed Cristina Cordova, former Partner at First Round Capital, as Chief Operating Officer. A credible executive joining an early-stage startup is always a huge plus, but in this case, it is an even stronger signal given the following context. Cristina is an A+ operator, having held senior positions at Notion and Stripe during their hyper-growth phases, and also has an asymmetric information advantage. As a former investor at a top-tier VC firm, she sees investor updates from her portfolio companies, gets exposure to new deals, and talks to other investors about startups (the job of a VC :)). When someone who seemingly any startup would love to hire picks a startup, you should take note.
Linear’s design culture starts at the top with its CEO, Karri Saarinen. Karri previously worked as Head of Design at Coinbase and Principal Designer at Airbnb. Linear’s product is visually appealing, intuitive to people working with product teams, and as simple as a Google Doc to contribute to. I know this from experience with the product since we used it at On Deck. With other product management software, I can’t say the same thing.
When I reviewed space startups focused on launching rockets, one thing that stood out to me was just how significant a lead SpaceX had regarding shipping speed and cadence. Since 2002, SpaceX has had 230 total launches, 191 total landings, and 165 total reflights. In comparison, Blue Origin, an incredibly well-funded space company bankrolled by Jeff Bezos, has just 22 launches since it was founded in 2000. The more shots you can put on goal, the more likely you’ll be successful. SpaceX is bringing in more revenue with each launch and learning from its missions.
While Elon Musk is considered a polarizing figure, it’s hard to bet against his business success. Elon is still one of the biggest talent magnets in the industry, which is highlighted throughout the SpaceX executive team. Many of its leaders joined early and have stayed their entire careers — such as Mark Juncosa (12 years), Bret Johnson (12 years), Charles Kuehmann (7 years), Joe Petrzelka (11 years), Will Heltsley (13 years) — a good signal when looking at a company from the outside. Elon has created an environment and mission for people to do their life’s work.
SpaceX is one of the most valuable private companies in the world. It’s currently valued at $137 billion. Even at that price, SpaceX ranked #1 in the Setter 30, a report on the most in-demand secondary shares based on a survey of 500 late-stage investors in Q4 2022. This marked the 8th consecutive quarter SpaceX was ranked in the top 2 alongside Stripe. At that stage, strong business fundamentals are necessary to pique the interest of investors since they will have to invest vast sums of money to make money on marginal price increases, unlike early-stage investing.
The space industry is still very new, and only in recent years have private companies begun to capture value that was previously monopolized by government agencies like NASA. Fortune reported that the global space economy was worth $546 billion in 2022 and is expected to grow 41% over the next 5 years. As the infrastructure continues to develop with new players commercializing other parts of the space industry, SpaceX has more opportunities to capture value on top of its rocket-launching and Starlink satellite Internet-providing business, which are strong revenue lines as standalone products. SpaceX expects to only capture $8 billion in revenue in 2023, so there’s still plenty of room for growth. We’re early, folks.
It’s much easier to iterate on a product early on than when your company has millions of dollars in funding and hundreds of employees. For starters, if you build something and no one buys it, it’s obvious that your idea is a dud, and you need to pivot. On the other hand, if you have paying customers and you’re growing, you must determine whether you’re growing fast enough to raise your next round of funding or reach profitability. Dalton Caldwell, a Partner at Y Combinator, describes this as a “weak signal trap,” one of the most dangerous things for a startup because it could cost the founders years clinging to a false sense of hope. If a startup is aware enough to realize they are getting a weak signal from the market, most founders and startups still end up folding. Sometimes it’s from an inability to make hard decisions — the founders may need to churn customers and fire loyal teammates; a failure to manage expectations with stakeholders — the founders may have sold investors one vision and need to pitch them a completely different vision; or a lack of motivation of the team and themselves — the founders and employees may be burnt out and unable to generate enough momentum to pivot the business. Each one of these decisions causes a chain reaction. For example, if a cofounder is burnt out and leaves, many employees who look up to that cofounder may also leave or blame the other founders for the exit and lose trust in their decisions. Brex is not one of those startups. When the executive team realized they were getting a weak signal from the market, they created a new product and refocused their team on a core customer, venture-backed startups. This meant shedding thousands of small business customers and even receiving negative backlash from the press. It was a big bet, but a necessary one to accelerate the startup’s growth once again and put them on track to grow into their valuation. Michael Tannenbaum, COO at Brex, shared that Pedro Franceschi, Founder of Brex, called this moment the “refounding of Brex.”
Brex is doing numbers. As of July 2023, Brex expects to hit $500 million in annualized revenue within the next 12 months. Brex counts 25% of all venture-backed startups as customers, totaling 20,000+ customers, including companies like Airbnb, Y Combinator, Carta, Classpass, Vouched, DoorDash, and Mutiny. For these customers, Brex manages billions of dollars in deposits, processes tens of billions in transactions, and is used in 100+ countries. This is what the 99th percentile looks like for a growth stage company.
Henrique Dubugras and Pedro Franceschi, the cofounders of Brex, have been building fintech startups together for over a decade. The two met over social media when they were 16 years old and started a company called Pagar.me, basically the Stripe of Brazil. The company had 150 employees and had processed $1.5 billion in transactions when they sold it and moved to the US to attend Stanford. They dropped out shortly after and attended YC with a completely different idea in the VR space before experiencing first-hand the issues with startups getting corporate cards. They launched Brex, and the rest is history. Like the founders of Ramp, Henrique and Pedro’s domain expertise and decade-long relationship helped them develop valuable insights and the operating rhythm needed to build Brex into a generational company.
By participating in Y Combinator (YC), Brex has accumulated quite the distribution advantage. It’s no secret that YC founders want to work with other YC founders. YC Partners support and even encourage founders in the cohort to sell to each other to get early traction and bring in the founders of successful startups like Stripe, Brex, and Rippling, who all target YC startups as customers, to speak to cohorts about what they’ve learned. As of July 2023, 80% of YC startups use Brex. Expect this number to continue to rise as 91% of YC’s latest batch, YC S23, picked Brex, indicating they are winning new cohorts over well-funded competitors like Ramp and Airbase.
My first full-time job was at a startup called Wefunder. Wefunder allows ordinary people to invest in startups for as little as $100. My boss at Wefunder was Nick Tommarello. I’d guess Nick has met 10,000+ founders in his career. That might be on the low side. Some people I know that Nick knows include Justin Kan (Founder of Twitch), Nathan Blecharczyk (Founder of Airbnb), Amjad Masad (Founder of Replit), and Immad Akhund (Founder of Mercury). When asked who is the best founder he knows, Nick said Parker Conrad, Founder of Rippling. Nick was batch mates with Parker during Y Combinator’s W13 batch. So he knows him well. As someone who respects Nick’s ability to evaluate talent (heck, he got it right by hiring me -- jk), I treat this as a strong co-sign (staking his reputation on).
The PayPal mafia is widely considered the most incredible collection of talent ever in tech — as employees would go on to found, invest in, and lead this generation’s most influential tech companies. What people don’t know is the Zenefits Mafia might be next. At Zenefits, the first startup founded by Parker Conrad, he assembled a world-class team. From one organization, Zenefits produced multiple founders who started companies valued at $1 billion, Partners at Founders Fund, Spark Capital, and Craft Ventures, and executives like the CRO at Brex, the VP of Talent at Databricks, the Director of Engineering at Salesforce, the CRO at Rippling, the SVP of Sales at Navan, the SVP of Sales at Gong, the SVP of Sales at Modern Health, the Head of Sales at Linear, and the Chief Customer Officer at Sendoso. I think the strongest signal that Rippling would work was that so many of the fantastic people he recruited for Zenefits followed him to Rippling, some immediately like Prasanna Sankar and Matt Plank. He continues to fill out the Rippling executive team with the best Zenefits alums, including Matt Epstein, Ashley Kelly, and Stephanie Ho.
As startups grow, one of the main challenges that leaders face is they get abstracted away from problems with each layer in their org chart. By the time an issue rises to the attention of a founder, it may have passed through 5 layers of people — the IC tells their manager, who then tells their manager, etc. — who each innocently alter the problem ever so slightly that by the time it gets to the founder, the problem is not what was first described. Parker sidesteps this problem by making himself, his engineering team, and every Rippling employee available to answer customer support so they understand first-hand what problems their customers face. He also knows that with a product like payroll, businesses care deeply about the level of customer support they get if something doesn’t work. Rippling is customer-obsessed, and they are not just saying that for marketing purposes. They walk the walk.
Garry Tan, President & CEO of Y Combinator, said Parker Conrad was a great founder cause he has a chip on his shoulder. Parker claims he always had this chip on his shoulder, but I’m sure he developed more of that mentality when he got kicked out of his first startup, Zenefits, amid a compliance scandal very publicly. To connect it with another thread when evaluating founders, Jessica Livingston, Founder of Y Combinator, says that determination is the most important trait of successful founders. Startups are incredibly hard, but I think Parker's chip on his shoulder makes him more determined than others and gives him an edge.
Speed of progress is a really good indicator to index for when evaluating a startup. Keith Rabois, Partner at Founders Fund, tweeted that Ramp has the fastest executing product and engineering team they’ve ever seen and followed it up later with this tweet about setting the bar for innovative velocity. For context, Founders Fund has invested in Stripe, Palantir, Anduril, SpaceX, Facebook, and Airbnb. You may be thinking they’re just marketing their book, but no. Founders Fund wouldn’t risk saying this about one investment if it wasn’t true because their brand is worth more than one startup in a portfolio of investments. This endorsement from Keith and Founders Fund speaks volumes.
As of July 2023, Ramp has 300,000+ users (~13,000 companies), processes $10 billion annually, and supports 195 countries and 40+ currencies. Customers include fast-growing startups like Deel, Faire, Attentive, Glossier, and Eight Sleep. Ramp also notes that between January 2023 and July 2023, Ramp’s customers with international operations increased by 2x. This is noteworthy because Ramp makes more money based on how much its customers spend from interchange fees. So Ramp’s growth is directly tied to its customers performing well and scaling operations. In March 2023, Ramp reported 4x revenue growth on $100 million+ in ARR, which it surpassed before March 2022. Startups between $75 million - $150 million in revenue are in the 75th percentile if they grow at ~65% YoY. So, based on this data, Ramp is growing at the 99th percentile for startups during this stage of growth.
Eric Glyman and Karim Atiyeh founded Ramp. Two things catch my attention immediately about this pair of founders. First, is that Eric and Karim have known each other for years. They met during college at Harvard University and launched their first company together during school called Paribus. Paribus did quite well — they saved their 700,000 users $100 million — before Capital One acquired it. There they noticed many problems with corporate cards and expense management software, which they would use as key insights to build Ramp. All of this is to say that Eric and Karim have been friends and cofounders for 10+ years. I’ve seen many investors allude to the time founders have known each other as a key indicator of future success. Second, Eric and Karim are fintech veterans. With access to years of qualitative observations and so much financial data from their time at Paribus, Capital One, and Ramp, they understand consumers better than almost anyone regarding finances and spending, which gives them an upper hand when building their product.
For a company that makes more money if its customers spend more money, it seems counterintuitive to build a product that incentivizes savings. But that’s just what Ramp has done from the very beginning. As of July 2023, Ramp’s saving insights, pricing intelligence, and expert negotiators have saved companies over $600 million and 8.5 million hours. This product decision helps build trust with its customers. Rather than winning on the edges in the short term, Ramp plays on a long horizon and wins based on the future growth of its customers. This is a beautiful example of a company aligning its interests with its customers.
The founders of Anduril are Brian Schimpf, Matt Grimm, and Palmer Luckey. Brian and Matt previously worked at Palantir, where the former was Director of Engineering, and the latter was a Forward Deployed Engineer. This is significant because Palantir sold directly to the government. Selling to the US government is notoriously hard, but Brian and Matt’s deep industry experience and existing relationships can help them position themselves better to win contracts. Palmer Luckey created Oculus VR. While the virtual reality industry has not taken off, Palmer should be applauded for his ability to build an incredibly technical product, something needed to create a defense tech company. Surrounding the founders at Anduril are the who’s who in both tech and government, including Christian Brose (Chief Strategy Officer), Matthew Steckman (CRO), and Scott Goldstein (Chief Scientist). This team’s combination of technical excellence and government relationships makes it an A+ executive team.
US government contracts are very competitive. Anduril has demonstrated an ability to win significant contracts from the US government and allies, such as this $967 million contract with the US and a $100 million contract with the Royal Australian Navy. Notably, Anduril is winning larger contracts more frequently than other well-funded defense technology startups like Shield AI and Epirus.
Only in recent years has investing in defense technology startups become more socially acceptable. Notably, Josh Wolfe from Lux Capital famously fought his firm’s Partners to invest in Anduril. That said, not every top-tier VC plays in the defense technology space. Even Sequoia Capital, widely regarded as the most prominent venture capital firm, just made their first investment in a defense technology company in June 2023 when they invested in Mach Industries. So, it’s impressive that Anduril has attracted investments from all of the top venture capitalists who have investments in this space, like Founders Fund, Andreessen Horowitz, and General Catalyst, even at growth-stage rounds and valuations.
Anduril’s vision is to create a family of systems powered by an AI-powered operating system, to transform defense capabilities. Rather than build the entire suite of products internally, Anduril is acquiring technologies to make its family of systems more compelling to governments. By acquiring defense technology, they get the technology and the very in-demand and specialized talent who built the product, while also de-risking their operation and accelerating their product timeline. They get a working product without having to invest heavily into R&D and other expenses tied to exploring and creating a product that may not work or have commercial value. In the last couple of years, Anduril has acquired Dive Technologies (autonomous underwater vehicles), Copious Imaging (passive sensing technology), and Aera-I (air-launched effects).
As of November 2021, Vercel’s Next.js web development framework gets 2 million weekly npm installs. Today, Vercel serves 50 billion+ weekly requests, up from 24 billion+ weekly requests in November 2021, from developers using its framework and guarantees a 99.99% uptime. Vercel counts large enterprises as customers, including Netflix, Adobe, Okta, HashiCorp, Zapier, Loom, TripAdvisor, Under Armour, and The Washington Post.
Guillermo Rauch is the Founder & CEO of Vercel. Many consider Guillermo an open-source legend. Guillermo created several popular Node.JS open-source libraries, including socket.io (58.7K stars), mongoose (25.9K stars), and slackin (6.5K stars). In addition, Guillermo started Cloudup, a file-sharing service with 10,000 users, which Automattic, the company behind Squarespace, acquired. While Guillermo’s previous success does not necessarily guarantee future success, it shows a deep understanding of developers and the ability to create products they want.
Jiaona Zhang, SVP of Product at Webflow and former product leader at Airbnb, WeWork, and Dropbox, popularized this concept of a minimum lovable product, contrary to YC’s minimum viable product advice. Jiaona argues that products must meet a certain quality bar, especially in a world where users have so many options. Vercel appears to operate with these principles in mind as they are constantly weighing the tradeoffs between wanting to iterate and not being afraid to make mistakes and ensuring production-grade infrastructure, as any downtime may result in millions in losses for their customers. Vercel’s product velocity is incredible, often shipping significant improvements and new products monthly, all while committing to a high-quality bar for the products they ship.
One issue with ambitious startups is that they want to solve every problem, often called shiny object syndrome. Vercel is being very intentional about saying no, and deprioritizing things that don’t align with their critical path to creating a large business. It’s my number one learning from reading this interview by Adam Nathan about how Vercel operates. Guillermo eludes to this several times, sometimes explicitly — building for frontend instead of frontend and backend, building a frontend cloud instead of an everything cloud, and building for React instead of every library.
This year, Vanta is filling out its C-suite with proven startup executives. They added David Eckstein as CFO and Jeremy Epling as CPO. David previously worked as CFO at Menlo Security, an $800 million company backed by General Catalyst, and Jeremey previously worked as VP of Product at GitHub, which was acquired for $7.5 billion by Microsoft. These are A+ hirings on paper and a notable signal to job seekers and investors that this talent level is willing to bet on Vanta’s future company performance.
In the current economic climate, where it’s harder to raise capital, startups are cutting costs and reducing their software spend to just products necessary for business survival. In January 2023, Vendr, a SaaS procurement platform with 500+ customers, reported that Vanta had the most net-new purchase requests on the platform. No other competitor, like Drata, Secureframe, or Thoropass, was listed in the top 20.
As of June 2023, Vanta reported ~$80 million in ARR and is adding ~600 customers per quarter. Andrew Reed, the Partner at Sequoia Capital who led the firm’s investment in Vanta, shared that Vanta sits in the top 1% of his investment portfolio in 5 key metrics, including revenue and customer growth. For context, some of Andrew’s active portfolio includes Figma, Front, Zapier, Loom, and Sourcegraph.
Usually, laws work against startups. But in this case, laws help drive demand for Vanta’s product. If your business operates a website like 71% of businesses in 2023, you probably need to comply with some compliance regulations. For example, the EU requires any business that collects personal information like user names to comply with GDPR. This applies to EU businesses and any business worldwide with EU visitors. So unless you’re restricting your website to a specific location, you’ll need to buy from a vendor like Vanta to remain GDPR compliant regardless of your business size or type, or risk getting fined.
Whether valid or not, people are scared, at an existential level, about AI. It’s a spectrum — some people are concerned that AI will take their jobs, others think AI will uproot entire economies, and some are concerned AI puts the future of humanity at risk. Anthropic understands this and is using it to their advantage. Of all the AI infrastructure startups, Anthropic has successfully branded itself as the one focused on safety. The headline on its website is “AI research and products that put safety at the frontier.” I count the word “safety” on their company page 7 times. They describe it as a science. In every funding (Series C) and partnership (Scale AI x Anthropic) announcement, the word “safety” is mentioned countless times. And the press believes them too — check out these headlines “A Radical Plan to Make AI Good, Not Evil” by Wired and “The $1 billion gamble to ensure AI doesn’t destroy humanity” by Vox. For good measure, they are also a registered Public Benefit Corporation, which means they are legally required to follow their charter to generate social and public good. Anthropic’s consistent branding is helping them build trust with consumers and enterprises, making them more likely to try their products.
For an experiment, I asked my friends, “If you could work at any startup, what would your top 5 choices be?” I learned a lot of smart people in my network are interested in working at Anthropic. They include a former Product Manager at Surge AI who writes a popular newsletter on AI startups, a friend who was Head of Operations at a Series B startup backed by Founders Fund, and a former senior software engineer at Airtable. If that isn’t enough of a signal for you to be interested, it appears that Spark Capital did the same thing before leading Anthropic’s Series C. Yasmin Razavi, the Partner at Spark Capital who led the investment in Anthropic, tweeted that Anthropic’s name always came up when they asked their network about the best teams working on the toughest problems in AI.
Anthropic’s founders are former OpenAI executives who left with the narrative that they wanted to build a safety-first AI company from the ground up. The founders include OpenAI’s former VP of Research, Policy Director, VP of Safety and Policy, and Technical Staff and Research Team members. There must be some merit to this because Anthropic’s founders are not the only people who left OpenAI due to concerns about AI safety. Elon Musk, a co-founder of OpenAI, has also mentioned AI safety concerns as it relates to OpenAI and how they are operating in numerous public interviews. Another interesting thing about Anthropic’s situation is the founders of Anthropic left millions on the table in the secondary market by leaving OpenAI to take the leap of faith to start a new AI company, one that competes directly with OpenAI. I love a team with conviction.
Like OpenAI’s partnership with Microsoft, Anthropic needs a partner to have the cloud computing infrastructure needed to power its research, products, and API services. Without it, you can’t train your AI models. Anthropic has that with Google, their preferred cloud provider. Google is a significant investor in the company. In addition to its partnership with Google, Anthropic has landed key deals with Scale AI and Zoom. Both partnerships give Anthropic more distribution by getting Claude, Anthropic’s AI assistant, in the hands of Scale AI and Zoom customers.
In March 2023, Silicon Valley Bank (the 16th largest bank in the US) collapsed, creating panic and sending shockwaves throughout the global financial system. Signature Bank (the 29th largest bank in the US) and Credit Suisse (the second largest bank in Switzerland) failed shortly after. Silicon Valley Bank had 40,000+ customers, primarily startups, and most switched to Mercury after the crisis. Since March 2023, Mercury has added 26,000 customers and is growing 2x faster than before the crisis. Mercury has become the new bank of choice for Silicon Valley startups.
Customers, and people in general, remember you more for what you do for them when they are at their lowest than when everything is good. The banking crisis was a terrifying moment for startups banking with Silicon Valley Bank — founders were unsure whether their deposits were safe and if they could make payroll next month. In a tight squeeze, many startups mobilized to help their customers, notably Mercury, Rippling, and Brex. Mercury expanded its FDIC insurance to $3 million, up from the industry standard of $250,000, and Immad Akhund, Founder & CEO, worked 12 days straight to help as many businesses as he could. Rippling raised money to help its customers make payroll if the FDIC didn’t guarantee deposits, and Brex offered its startup customers an emergency credit line to help them make payroll. In a time of crisis, Mercury earned the trust of customers, and the data backs it up — 95% of customers who previously banked with Silicon Valley Bank have stayed with Mercury 90 days in.
Immad Akhund is the Founder & CEO of Mercury. Immad previously founded Heyzap, a mobile ad network to help users discover apps they love. Heyzap raised $8 million from Union Square Ventures, Y Combinator, Naval Ravikant, and Chris Dixon before being acquired for $45 million by Fyber. While building companies, Immad has built a reputation as scrappy, thoughtful, customer-centric, and a big thinker. He constantly challenges his team to think on long horizons and 10x their ideas. Mercury is Immad’s life’s work.
Unlike traditional banks, Mercury leverages partner banks and sweep networks to distribute risk across a network rather than have a single point of failure. That means when you bank with Mercury, they spread your deposit across a network of FDIC-insured program banks. This allows Mercury to maximize FDIC insurance for its customers without the manual work of opening and moving money across separate bank entities. By protecting their customers from financial risk, they also protect themselves.
Manufacturing, specifically space manufacturing, is reaching a tipping point in the US. The current system is standing on thin ice as startups like SpaceX and Blue Origin rely on near-retired machinists, working out of mom-and-pop shops across the country, to deliver quality grade parts on time to launch rockets. It’s inefficient, unreliable, bad for their bottom line, and potentially dangerous. Hadrian is in a perfect position to meet the unmet demand from SpaceX and Blue Origin and newcomers like Relativity Space, ABL Space Systems, and Apex. As of July 2023, Hadrian can deliver flight-grade parts in 5-21 days, 10x faster and more efficiently than anyone else.
In 2021, Elon Musk tweeted the factory is the product. It’s a pretty famous tweet because Telsa is known for productizing their factories with technology, automation, and systems to produce Teslas as fast and efficiently as possible to deliver on demand. This type of architecture gives them a compounding advantage compared to other car manufacturers since it can support massive scale. Hadrian can be the factory for the growing number of space startups in the US. As of 2021, there are 5,582 space companies in the US, and the industry is expected to grow 41% over the next five years. Hadrian’s product is so good they are hiring former copywriters and baristas who can produce parts in their factories within 30 days of onboarding more efficiently and quicker than machinists with a lifetime of training.
One of my friends, whose opinion I hold very highly, picked Hadrian as one of her top choices when asked about the startups she’d be excited to work for. She texted, “The founder is excellent, and I believe its [mission] is important.” She’s a former Investor at Bloomberg Beta. Another one of my former colleagues, early-career but very ambitious, also picked Hadrian over other exciting startup and investing opportunities.
In March 2023, Ben Braverman joined Hadrian as Chief Business Officer. Ben was an early employee at Flexport and instrumental to the company’s success. During his ~9 years at Flexport, Ben built and scaled Flexport’s go-to-market and corporate development teams. Out of all the executive moves I’ve tracked this year, Hadrian landing Ben is top 5 in terms of significantly impacting a startup's chances of success. Along with his expertise, Ben will raise the hiring bar and help Hadrian navigate growth-stage challenges that he’s already seen once at Flexport. Only a few months in, Ben landed a big win by expanding their partnership with Anduril.
Several market tailwinds have increased the need for Anrok’s enterprise-grade sales tax solution. In 2018, the South Dakota v. Wayfair Supreme Court decision ruled that states may charge tax on purchases made from out-of-state sellers. As a result, if sales tax is not collected from the customer upfront, it comes out of the company’s pocket. Michelle, Anrok’s CEO, shared that sales tax may impact a company’s bottom line by 4.3%-11%, depending on which states the company is selling to. Oh yeah, if they don’t do it, the company accrues interest and penalties and is subject to being audited by several states in a given year. This law change, combined with data showing most SaaS startups are selling to customers in over a dozen states, has increased demand for Anrok’s solution. Another market tailwind is the rise in remote work. By 2025, Upwork estimates that 32.6 million Americans will be working remotely, or 22% of the workforce. A company employing a remote employee creates an immediate sales tax obligation.
Michelle Valentine and Kannan Goundan met at Airtable as early employees on the Product and Engineering teams, respectively, and started Anrok after realizing finance teams were worried about what to do with sales tax. Before that, Michelle was an investor at Index Ventures and an Investment Banking Analyst at Goldman Sachs in the public sector infrastructure group. While Anrok is still a small team, they’ve been able to recruit a really impressive early team, including Brad Silicani (former Treasurer at Dropbox) as COO, Devon Watts (former VP of Marketing at Groundspeed Analytics) as Head of Marketing, and Eleanor Henderson (former AWS Tax Planning at Amazon) as Head of Tax.
As of May 2022, Anrok reported a 7x increase in revenue in the last six months. Anrok’s sweet spot for customers is hyper-growth startups. They count Notion, Vanta, Pave, Front, and AgentSync as customers. From Anrok’s case studies on its website, it’s easy to see how much money they are saving customers (AgentSync was hit with a $500,000 tax bill in 2021) and how quick the solution is to set up (Pave set up on Anrok in 30 minutes). It’s also evident that Anrok can support complex edge cases for customers, which Michelle confirmed in interviews, due to their focus being on SaaS companies vs. incumbents like Avalara, who supports a larger mix of business types.
Anrok is backed by Sequoia Capital and Index Ventures. Both VC firms co-led the $4.3 million seed round and $20 million Series A round. This is a strong signal as VCs want to maximize their ownership in the best investments in their portfolio. There would also be signaling risk if Index Ventures never participated, as Michelle was a former Investor at the firm.
Andrew Reed, Partner at Sequoia Capital, tweeted that Warp has one of the most talented engineering teams. John Rector leads Warp’s engineering team. John previously was Senior Director of Engineering at Reddit and Co-Founder & VP of Engineering of Dialpad ($2.2 billion valuation). John’s team is now 25+, including software engineers from Facebook, Dropbox, LinkedIn, and Youtube.
Zach Lloyd is the Founder & CEO of Warp. Zach previously founded SelfMade, a startup bringing professional editing to your phone for pictures. Founded in 2015, SelfMade raised $11 million in venture capital from top VCs, including FirstMark, GGV Capital, and SV Angel. Before that, Zach had a 7-year stint as Principal Engineer at Google on the Google Sheets and Google Docs Suite team. During that time, Zach’s team re-wrote the entire Google Sheets application to add offline editing capabilities and other performance improvements that led to a 40x increase in usage. At large companies, the closer an engineer is to the core product, the better the engineer. Like Zach, an engineer working on Google Sheets is likely more impressive than an engineer working on Google+. While Zach’s engineering experience makes him a perfect fit to lead Warp, I think the strongest signal is his obsession and curiosity about the problem. In his interview with Fast Company, Zach shared that he spent way more time than the average geek thinking about the command line and how to improve it without stripping away the core principles that have made it relevant for so long.
Engineers are really expensive. The median salary for an engineer at OpenAI is $925,000 in total compensation (base salary + stock). With the significant increases in compensation for engineers, startups are looking for ways to increase engineering productivity and extend and retain their engineering teams. For example, Stripe acquired Okay to help their engineering managers better understand their teams’ effectiveness and drive more efficient performance. The entire point of the acquisition was to put themselves in a better place to attract and retain talented engineers. I expect more startups to prioritize this, so Warp’s bet that companies will pay for their developers to use the product seems like a good one.
Warp launched with mixed reviews on Hacker News in 2022. While some developers raised valid concerns about security and privacy, Warp used that feedback to incorporate them as foundational elements of its product. With its commitment to improving the terminal, Warp has received endorsements and investments from Sam Altman (OpenAI), Dylan Field (Figma), and Marc Benioff (Salesforce), all founders with programming backgrounds. Dylan said, “It’s a really, really good product if you’re a developer.”
As of July 2023, Pulley has 3,000+ customers, up from 1,700 in July 2022, including Coda, Y Combinator, Branch, Varda, and Athelas. Pulley’s growth appears to be accelerating as they added 600 customers in 2020 and 1,300+ between July 2022 and July 2023. Pulley is the top-rated cap table solution according to G2 reviews, ahead of Carta, Shareworks, and LTSE Equity. It also seems to be the cap table of choice for top accelerators as it counts Y Combinator, Techstars, and 500 Global as customers on its homepage. In 2020, Pulley reported over 50% of YC startups picked them as their cap table solution.
As of July 2023, Pulley reported that their average response time is less than 5 minutes, and in their G2 reviews, many customers have applauded Pulley’s excellent customer service. On the other hand, Carta, Pulley’s biggest competitor, has received several complaints about poor customer service and their lack of transparency regarding pricing. Funny enough, Carta’s pricing page confirms this. They don’t have a single price listed on it. Instead, you have to talk to a sales rep. The comparison between Pulley vs. Carta shows they are significantly faster in just about everything, onboarding, 409A valuations, stock splits, etc.
Similar to Carta’s suite of products, Pulley has room to build new products around equity management that are valuable and organic to their current customers. As of July 2023, some one-off solutions include waterfall modeling and custom reports. Other products could vary from compensation benchmarks to offer letters. As Pulley’s suite of products expands, they can cross-sell different packages to customers based on triggers and data they are collecting on the startup. If done well, Pulley could drive more incremental revenue from each customer and offer a more valuable product.
Pulley has put together a really strong team to solve problems around equity management. Yin Wu, the Founder & CEO of Pulley, is a repeat founder and talented engineer. Pulley is her 4th startup. All have been venture-backed, including Oven Labs (backed by Andreessen Horowitz), Echo (Acquired by Microsoft), and DIRT Protocol (backed by SV Angel). For a startup positioned as wanting to solve equity management for founders since they believe founders drive better startup outcomes, Yin, someone who’s experienced these problems firsthand for her previous startups, is the perfect person to lead this team. Yin and her cofounder, Mark Erdmann, have assembled a stellar team, including Jason Lee (Former Director of Sales at Flexport) to lead the Revenue team, Dan Becker (Former Head of Product Design at Instagram) to lead the Design team, and Ti Zhao (Former Founder of Kip, acquired by Modern Health) to own Product decisions.
Since releasing V1 of its product, The Browser Company has seen exponential growth. As of July 2023, The Browser Company has 100,000+ users, up from 751 in June 2022. The growth rate is still steep as they reported 2x active member growth in Q2 2023 over Q1 2023. The thing is, I expect The Browser Company’s growth to get even better. In an update on Youtube, The Browser Company reported losing 80/100 customers because of not being available on Windows and their waitlist – two things ~75% of their team is now focused on fixing. In less than three weeks, The Browser Company revamped its website, removed its waitlist, and optimized its onboarding flow. As someone who went through the new flow, I think the fresh website design and onboarding flow is simple but effective.
In 2021, Google Chrome, the most popular browser, had 3.2 billion users, and Safari, the second most popular browser, had 576 million users. Google doesn’t share revenue numbers for Google Chrome but has communicated it is “extremely profitable,” and many analysts believe it has added billions to the bottom line. The Browser Company is playing in a huge market that will continue to expand. We take it for granted, but many places still don’t have access to the Internet, an underreported problem since economic growth is so dependent on the availability of it. As of April 2023, Statista estimated only 64.6% of the global population, 5.18 billion people, have access to the Internet.
While still an early-stage company, having raised under $20 million in capital, The Browser Company has assembled an exceptional engineering team. Hursh Agrawal, former CTO at Thirty Madison ($1 billion+ valuation), and Dolapo Falola, former Director of Engineering at Slack, lead The Browser Company’s engineering team. The two have recruited talented engineers from Instagram, Google, Netflix, and Apple. This type of talent is what they need to have a chance to deliver on their mission of creating the best Internet browser.
Google Chrome and other browsers indirectly make money by selling user data to advertisers. The Browser Company wants to do the exact opposite, electing to monetize through Arc for Teams (give it away to individuals for free and charge them for Arc if they use it together as a team — similar to how Airtable, Notion, and Figma have monetized) and Boosts (a community store or marketplace where people can use other cool things that people have built-in Arc). The Browser Company’s monetization strategy may help them in the long run as more people are concerned about sharing private data with companies.
Sometimes, having co-founders not involved in the business’s day-to-day operations can be a negative signal. Startups are really hard, and if the founders are hedging their bets because they want to dip their hands into many ventures, then the startup may never reach the success it otherwise could have with full focus. The way that I look at it is two-fold, the caliber of the founders that are doing this and whether it is clearly defined. I view Elon Musk’s founder ability as a benchmark for this. As Founder and CEO, he’s worked across multiple successful startups — Tesla, SpaceX, and The Boring Company. If you’re not even in the same room as him, then it’s probably a negative signal. Brian Armstrong and Blake Byers are not just in the room, but they have a seat at the table with him. Brian is the Founder & CEO of Coinbase, the category winner in crypto, and Blake Byers is a former Partner at GV. From what I can see from the outside, I think Blake will play a more critical role in this company. He has the technical knowledge, a Ph.D. in Bioengineering at Stanford University, and investments in biotech, Arcus Biosciences, Benchling, etc., to have developed a valuable network to support this company. You can see him actively recruiting for NewLimit on his LinkedIn. He also probably has more time than Brain, who is still CEO at Coinbase and fighting with the SEC. On the second point, both of their roles with the startup have been clearly defined from the start and is publicly available information as they have mentioned it in the press.
I am not a biotech expert. In situations like this, it’s best to listen and learn from experts, such as Celine Halioua, in the space. Celine is also working on solving human aging, starting with dogs. Her company, Loyal, just received FDA approval on their companion dog longevity study and is well funded by Khosla Ventures, First Round Capital, and The Longevity Fund. Here’s the interaction between Celine and Blake Byers, one of the co-founders of New Limit. While this is not an endorsement from Celine, I view it as a positive signal indicating that she believes NewLimit has developed an approach that seems viable and worth the risk/reward for patients.
Unlike software, when you look at biotech startups, it’s tough to gauge speed of progress. Usually, there is a graphic with a drug pipeline, which shows you where the drug is at in development. However, there are no labels indicating how long the drug has been in that phase, how long it’ll take to advance to the next step, what needs to happen next to move to the next stage, what risks could block the drug from getting to market, etc. I imagine they don’t share these things because the timeline is also fuzzy internally, and many biotech startups don’t move fast as they face regulatory, hiring, and technical challenges. So from the outside, it’s almost impossible to tell whether they are moving fast or are behind from the vague drug pipeline graphic and sporadic blog posts. What I like about NewLimit is you can see what they accomplish every month. They publish monthly progress updates and make it easy to understand, so someone like me, who doesn’t know much about biotech, can follow along. I think it’s impressive that they are moving fast enough to share monthly updates. I know software startups who wouldn’t want to do that, and I think the transparency will help them recruit talented people who want to join a biotech startup with momentum they can see.
In May 2023, NewLimit closed its Series A from Founders Fund, Elad Gil, Kleiner Perkins, Dimension Capital, and Garry Tan. With science-based investments, the due diligence process is often more thorough. Unlike software investments, VCs need to evaluate the scientific difficulty and the startup’s ability to scale it economically, on top of standard evaluation criteria focused on the team, market size, etc. Also, when you invest in a software startup that fails, you lose the money and go next. When you invest in a biotech or healthcare startup that fails because the underlying science behind the startup is wrong, you lose the money and look like an idiot. Heck, they might make a Netflix documentary out of it. So, I think there’s more reputational skin in the game with science-based investments than software investments, which matters much more than money to firms like Founders Fund and Kleiner Perkins.
Businesses are more cross-functional than ever, so cross-functional leaders need financial data to make decisions. At On Deck, I often had to request data from our VP of Finance or stitch it together from Stripe, Slack notifications, and random Excel files. It was messy, tedious, and a huge time suck. In a non-zero interest rate world where margins, quality of revenue, and burn rate matter, Runway Financial gives finance teams the superpowers needed to provide visibility into the org to the right people at the right time.
Siqi Chen, a repeat founder who previously founded Sandbox VR, is the Founder & CEO of Runway Financial. Sandbox VR is a Series B company backed by Andreessen Horowitz, Alibaba, and Craft Ventures, with a very complicated structure that involves hardware, games, content, virtual reality technology, retail stores, and more. While building that company, Siqi saw firsthand how painful it was to build financial models from scratch and the problems created due to siloed financial data across different teams. Runway Financial’s vision is to become the “Figma for Finance,” and Siqi is a formidable founder to attempt to reach this vision.
As of July 2023, Runway Financial is growing at more than 10x YoY, with a pipeline booked until the end of the year. That was before all the new customers they will get from their Series A announcement. What’s more impressive to me is that Runway Financial is committed to building a product that customers actually love. Siqi shared that they even scraped their first product because it demoed well but wasn’t valuable enough. A commitment to product excellence and crafting something people will love will only improve customer retention and help ensure they don’t overestimate their product-market-fit, a fatal flaw for many startups. This approach seems to be working as Runway Financial’s new product has received flattering reviews from people like Sandy Cass, former CFO at Artsy and VP of Finance at Bonobos, and Sasha Orloff, former VP at Citi.
Runway Financial is the only financial platform in the world with a Rippling integration. This improves Runway Financial’s product because it allows Rippling customers to sync employee information like compensation data into their dashboard directly, and the company’s model, forecasts, and reports will always remain up-to-date and accurate. In addition, the Rippling partnership will help Runway Financial get more distribution. Rippling customers will be inclined to try and want to use the recommended product from the quasi-gated Rippling app store.
The pictures look like the photos. One of the biggest problems with Airbnb or other short-term rental platform like Facebook marketplace is booking something that turns out to be different from the pictures. I remember getting scammed on fake and misleading pictures when I tried to rent a place for my 3rd year of college and later on several Airbnb trips. It’s the type of experience that makes you want to stay in hotels because at least you know you’ll get a clean bed and towels, and they’ll clean up after you. Wander controls for this by owning and maintaining the inventory. Customers get the peace of mind that when they book at any property on Wander, they get an up-scale and consistent travel experience.
After learning about Wander in January 2022, I started following their Founder & CEO, John Andrew Entwistle, on Twitter. I’m not too active on Twitter, but I remember seeing an interaction between John and a customer on one of his tweets. John was teasing a new product, and a customer mentioned they were staying in a Wander next month. John immediately told the customer to DM him the email under which the reservation was made so the team could do something special. Not only does Wander have excellent customer service to any complaints, the standard for any travel company, but they are also taking steps to delight customers. You can read the entire interaction here. It wasn’t even a high-profile customer — he has 9 followers! John is leading by example, and the Wander team has no excuses but to follow.
Wander is more than just a short-term rental home. Instead, Wander does everything it can to curate a full-service travel experience by adding amenities, content, and taste to the trip. Think Teslas so you don’t have to book a rental car, recommendations so you don’t have to scroll through dozens of affiliate travel articles, and additional amenities like beach essentials for ocean-front properties and private hiking trails for Wander cabins. The exceptional real estate is just one piece of each customer’s planned travel experience.
When I first learned about Wander, they had one property. I remember downloading the app and seeing if I could book a trip. The first spot was six months+ out! Fast forward to now, and Wander has 16 properties. Still, each property has reservations for the next 2-3 months. Wander has 175,000+ users, crossed 5,000+ nights booked on the platform, and boasts a 94.6% customer satisfaction rating.
It blew my mind when I heard the Lipstick Brother sell $1.9 billion worth of products in one live stream with Alibaba. I’m seeing more reports, like this one, that show that consumers are influenced by creators and embracing it. I predict in a few years that, the data will continue to progress to the point where consumers trust their favorite creators over legacy brands and want to buy products aligned with them, especially when the attribution issues get fixed.
The biggest creators realize there is more money to be made by creating and selling their products than getting paid to advertise someone else’s. Look at Kim Kardashian with Skims, Logan Paul and KSI with Prime Hydration, and Tom Brady with Autograph. If Flagship can lower the barrier to entry for creators to start their own companies and sell products, then more creators will choose to monetize this way rather than traditional ads and brand deals. You’ll begin to see mid-level creators, not just the Kim Kardashians of the world, generating enough income through their products to go full-time.
I was early in influencer marketing with my twin brother. In 2013, we created an Instagram account that grew to 2 million followers in a few years. We tried to monetize the account but found it incredibly difficult since many of our brand deals were one-off posts which hindered our ability to bring in revenue consistently. It felt like we were running on a hamster wheel — we were always looking for the next deal, not knowing where it would come from. Fast forward to 2023, and the same problems we faced back then are still prevalent. Creators are looking for new ways to monetize, and if Flagship can help them capitalize on their influence, they will become a market leader.
The founders of Flagship, Youssef Ahres, Khalil Hajji, and Juhana Kangaspunta, have uncanny founder-market fit. Youssef was the Director of Data Science at Instagram, where he built tools to help businesses succeed and worked directly with advertisers. Khalil previously worked on making YouTube recommendations better to drive growth for top creators on the platform. Juhana fills in the gaps for the team, providing them an exceptional technical talent who built things at places like Google and Zoox. Their founder-market fit and complementary skillsets make them an incredible founding team to tackle this problem.
Michael Chow and Steven Snow are the founders of Believer Entertainment. Michael previously was a VP and Executive Producer at Riot Games, where he worked on League of Legends and Wild Rift, the modified mobile game based on League of Legends. Before that, he founded two game studios. The first game studio, NewToy, created Words with Friends, the number one gaming app between 2010-2011, and was acquired by Zynga. The second game studio, Jolly Company, was acquired by Riot Games. Steven previously was an Executive Producer at Riot Games for League of Legends and an Executive Producer at Perpetual Entertainment for Star Trek Online. After stints working on some of the most memorable games, Steven also created his own gaming studio, Ministry of Hijinks. Michael and Steven make a formidable founding team as they have accomplished incredible things in the gaming industry at various vantage points — as founders designing and iterating on the first play to seasoned executives at large gaming companies scaling games into billions in revenue.
The best executive team in gaming that I’ve seen is at Believer Entertainment. When an early-stage startup has filled out its C-suite and VP level already, it’s usually a bad sign since there might be title inflation, but even worse, one subpar executive can result in 5+ hires that don’t meet the bar. When this happens, it can erode the quality of the products and the team’s ability to deliver to customers. The leadership team at Believer Entertainment includes executives, all of whom were executives at top gaming companies, like Landon McDowell (Riot Games), Jeremy Vanhoozer (Bungie), Tim Hsu (Riot Games), Grace Park (Riot Games), and Jeff Jew (Riot Games). This team has built games, scaled games, and monetized games. If you like gaming and want to bet on a team, this is the one.
I don’t game much anymore, but I still consume gaming content. One thing I learned recently, and obvious to me in hindsight, is that it’s much better to create the game than to operate in other parts of the gaming industry. For example, 100 Thieves, a competitive e-sports team funded by Sequoia Capital, lost $2.5 million on their Call of Duty team that won a World Championship. In July 2023, they decided to part ways with every championship team member and take on new cost-effective talent. While they hope to remain competitive, 100 Thieves learned that they couldn’t generate profits, even from a Championship-winning team in a very popular game. In a podcast episode, they shared that they are now experimenting with other business models to see if they can find a profit center. One of those bets is making their own games, called internally Project X. If you look at the top public gaming companies by market cap, they all make games. Microsoft has Minecraft, EA has the FIFA series, and Activision Blizzard has World of Warcraft. Much of the Believer Entertainment team has worked on one of the most successful games ever, League of Legends, which brought in $1.8 billion in revenue in 2022. They understand how to monetize games, and because they are creating the IP for the game, they are operating in the part of the industry where if they create a good game, it can be a profit center.
The studio, which had only 12 employees in March 2023, just raised $55 million from Lightspeed Venture Partners and Andreessen Horowitz. While both firms are most well-known for their technology investments, they have excellent track records in gaming. Lightspeed Venture Partners invested in Epic Games ($32 billion valuation) and Faraway Games (Series A, 400,000+ monthly active users). Andreessen Horowitz invested in Zynga ($9.30 billion market cap), Roblox ($24.22 billion market cap), RTFKT (acquired by Nike), lowkey (acquired by Niantic), and Parsec (acquired by Unity Technologies). Riot Games also invested in the round. It’s always a good sign when your former employer backs you.
LangChain’s open-source chatbot on GitHub is one of the most popular products for developers by all accounts. In ~8 months, LangChain’s repository has 56.5K stars, 7.4K forks, 15.3K users, and 1.2K contributors. Having looked at many developer tools startups and their open-source traction, I’ve never seen growth like this. Not only is this growth rate an accomplishment but reaching 25,000+ stars is a milestone in itself. As of July 2023, there were only 698 repositories that have reached that mark. While incredibly fast traction does not always indicate future success, especially in the world of AI where some startups appear to have no competitive moat, see Jasper and their layoffs shortly after reaching $75 million in revenue in 18 months and a unicorn valuation, it is worth keeping an eye on whether LangChain can maintain its lead at this layer in the AI stack relative to new competitors in the market.
LangChain is backed by Benchmark and Sequoia Capital. The funding history caught my attention because one week after Benchmark led a $10 million round in LangChain, Sequoia Capital led a $20 million round in LangChain. It shows that Sequoia Capital has a lot of conviction in the company, and there is a ton of investor interest around LangChain. They either led that round because they thought if they waited longer, another venture capital firm would take it from them, or they had so much conviction in LangChain that they were willing to invest at a higher price than Benchmark without waiting for more information and progress. It was likely a combination of both.
In July 2023, LangChain landed Brie Wolfson as a Marketing hire. This is a high-signal executive move because of two things — any startup would be lucky to hire her, and she has more information than anyone else. Brie previously created Stripe Press and Figma Education and consulted with high-growth startups like Replit, Ramp, and On Deck on culture-building. That’s how I first heard about Brie — she was at some On Deck events & team meetings. More recently, Brie started a company called Constellate that Sequoia Capital funded through its Arc program. After shutting the company down, I’m positive Sequoia Capital shared the best opportunities in their portfolio with her, typical of any VC firm with a founder they’ve backed looking for their next thing, and she picked LangChain. She also had the pick of the litter from startups outside of Sequoia Capital’s portfolio.
While LangChain is powerful as a standalone solution to help developers build chat-based applications on top of LLMs like ChatGPT, it becomes more powerful when combined with other developer frameworks. In May 2023, Microsoft’s CTO, Kevin Scott, announced in a keynote that their new tool, Prompt Flow, unifies LangChain and Semantic Kernel. LangChain added support for Cloudflare Workers, Vercel/Next.js, Deno, and Supabase Edge Functions a month earlier. Look for LangChain to continue adding more support with developer frameworks, and with it will come new and exciting workflows and product experiences.
It is no surprise to anyone who has tried to book travel accommodations or a hotel in the last couple of months that travel prices are rising. Kindred offers a no-brainer deal for travelers as the price to travel on Kindred is ~5x cheaper than booking an Airbnb or hotel.
As of July 2023, Kindred has accepted 4,500 members in 50 cities on the platform. They also have a waitlist of 25,000 people. As long as Kindred can maintain quality as the network continues to expand, the community will become more valuable as members can visit more destinations and have more options to choose from in each destination.
Companies like Airbnb and Uber were founded on innocent sharing economy principles, people helping people have a place to stay and a hot breakfast, and carpooling getting to work together. Unfortunately, as the platforms scaled, the sharing economy principles eroded due to bad actors and people looking to take advantage of the marketplace. Airbnb has dealt with things like Airbnb arbitrage, and Uber has been branded, maybe incorrectly so, as a platform where a small number of full-time drivers complete most of the rides. Kindred understands how harmful this can be to a platform and has taken several proactive steps to guard against this. One of these steps is that Kindred ensures members list only their primary residencies on the platform. This helps manage the quality issues, as primary residences are more suitable for living in since you notice things in your home that work well or don’t work well and fix them. In contrast, secondary residencies are often treated with less attention as you may only stay there for a couple of weeks out of the year and can put up with it. Another thing that Kindred implemented is a credit system. The more nights you host, the more credits you can travel with—a 1-for-1 swap. By using credits, Kindred incentivizes people to be active community members; what you give is what you’ll get. It also protects against bad actors who join with the intention of making extra money as a side hustle. Kindred is creating a community with healthy sharing economy dynamics.
Since I first heard about Kindred through the On Deck community in October 2021, Kindred’s hiring momentum has accelerated. As two non-technical founders, Justine and Tasneem, the most critical hire was landing Bryan Li as CTO in November 2022. Bryan was previously Director of Engineering at LinkedIn and Twitter and started a healthcare company called Medmonk, which Y Combinator funded. In addition to Bryan, Kindred has landed new hires from target companies like DoorDash, Faire, Airbnb, Tinder, and Bain & Company. With continued growth, I expect Kindred to begin to fill out its VP and Director levels soon.
At On Deck, where I worked for 2.5 years, the number one conversation at the company was how to ensure network quality when our entire business model depended on making revenue from accepting applicants into the community. It’s two directly competing incentives. Generally speaking, the more people you add to a community, the less exclusive it feels, even if the caliber of people joining is the same. Other community businesses like Chief are running into similar issues as they scale. Buildspace side-steps this network quality vs. revenue conversation in two ways. First, they partner with companies, funds, and governments to fund programs rather than relying on applicants to pay out of pocket. Additionally, quality is completely centered around someone’s ability to build instead of work and university credentials (i.e., PM at Facebook and Stanford University graduate). With these expectations set, Buildspace has a better chance of scaling its community more seamlessly without worrying as much about the perception of its network going down as they scale.
So many people have been funneled and brainwashed into a world where SaaS is considered the only viable business model. As valuation multiples on public SaaS companies have decreased and more people are getting into entrepreneurship from unconventional backgrounds, not just the Ivy League to Y Combinator pipeline, we’re seeing an increase in the number of people building stuff they are passionate about. For example, indie hackers like levels.io and dannypostmaa have built large online followings off of the success of multiple bootstrapped companies. I think people are more open to this path of entrepreneurship because it seems more attainable to be a successful indie hacker than the next Mark Zuckerberg. You can build literally anything that you’re passionate about at Buildspace, even a font!
Buildspace’s marketing is refreshingly authentic. For a quirky brand, they use skits, memes, and references to pop culture to create content that appeals to their audience. Some of my favorite skits, which are actually funny, are DMW-069 and apply with any idea.
Farza Majeed is the Founder & CEO of Buildspace. On paper, Farza looks like a typical venture-backed founder, a former deep-learning engineer at Visor.gg (acquired by Niantic), before getting accepted into Y combinator, the most prestigious accelerator in the world. But underneath the hood, when you really look into Farza’s path, you start to understand why he is the perfect person to lead Buildspace — a place where people build cool shit. From selling blank DVDs on eBay at 13 to making Youtube videos every day when he was 18, Farza is a builder and a damn good one at that.
