Going Solo

Mike Annunziata spent most of a decade on the LP side before writing venture cheques of his own. He began at the Soros family office as the note-taker in its private equity pipeline meetings before moving on to the Cornell endowment, where he covered every asset class and learned where the bar sits for an institutional manager.Mike is now Founder and Managing Partner of Also Capital, a concentrated hard tech fund. It grew out of the angel investing he did with his college friend, Will Bruey, now CEO of Varda. Their early cheques went into Varda and Radiant, and Also later wrote the first cheque into Northwood Space ahead of its Founders Fund seed round.In this episode of Going Solo, we discussed:* An emerging manager has to clear an LP’s opportunity cost. For an endowment like Cornell, the alternative is a marginal dollar into Andreessen Horowitz or Sequoia. A new fund needs a specific path to beating that by enough to cover the extra risk, and Mike doesn’t think “outcomes are bigger” justifies much of the portfolio construction he sees.* Also asks founders to assume their technology works, then answers around ten questions about enterprise value and how the company will capture it. Expert calls on lasers, nuclear, space or robotics come last. They serve as a gut check on whether the founder is taking the right degree of engineering risk.* Origination means lending credibility to founders who are less legible to the venture establishment. Bridgit Mendler hadn’t started a venture-backed company before Northwood. Also agreed with the ground station problem she was solving, and so did its SpaceX advisers. With that behind it, Mike’s introduction to Founders Fund moved the pitch to “where do I wire?”* Fundraising is selling a luxury good: nobody needs another fund, so they have to want it. Fund 1 took Also a year to raise and Fund 2 about ten weeks. Mike’s advice is to get into business and put money to work before the full target is raised.We also got into what Dorm Room Fund took from First Round’s process for controlling groupthink, and why Also backed Atlas on “allied production” while everyone was talking about domestic manufacturing. We finished on why Mike’s picture of the next ten years is robots everywhere.Hope you enjoy it.DG This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit blog.joinodin.com

Yohei Nakajima came to venture as a builder. He has made tools as a hobby since high school without ever holding an engineering job, and he began sharing them during COVID. A lot of no-code deal flow followed, then Web3 deal flow after he sold his own NFT art, and now he does it deliberately with AI tools.Yohei is General Partner and co-founder of Untapped Capital, a small firm on the frontier of applying AI to venture capital. His route there began about ten years ago at TechStars, helping spin up the Disney Accelerator, and runs back to a college thesis on angel investment organisations.In this episode of Going Solo, we discussed:* Building in public is a marketing strategy a VC can repeat for decades. Most VCs end up marketing a history of what they’ve done, and that only lasts so long. Sharing no-code tools brought Yohei no-code deal flow, selling his own NFT art brought Web3 deal flow, and with AI that approach became intentional.* Yohei explains why he starts from the assumption that venture could be fully automated and then the parts a firm keeps human says a lot about their uniqueness, with decision-making scoring the last thing Untapped plans to automate.* “Is this a good investment?” is a horrible question to ask AI. Yohei asks for the hypotheses that would make an investment more or less successful, plus the existing proof points for each, so the decision stays with the human. Automatically filtering out non-fits is a good use of AI, while fuzzy ranking of possible fits is the trap, because everybody ends up looking at the same things.* Early-stage investing is a discovery game. In the game behind Yohei’s shared discovery paradox paper, players do best when they share clues and spread their guesses across different boxes, because players who do not coordinate all pick the same box and lower the group’s chance of finding the prize. Broadly, alpha comes from discovery and beta from estimation, which is why venture needs many small firms alongside the large ones, and Yohei is extending the research through an autonomous research lab called Distributed Discovery.* Pitches have never been more polished, and a lot of it is AI, so reviewing unpolished signal matters more today. Yohei reads founders’ X reply tabs and the posts that drew engagement for the raw interactions that give personality signals, and he prefers a one to two page memo as an initial pitch.We also got into whether a fully automated VC trading secondaries is possible (Yohei thinks it is), two ways to build an AI-ready firm, and what stays the same in ten years: human connection, family and education.Hope you enjoy it.DG This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit blog.joinodin.com

Jenny Xiao came to venture from research. Following a PhD at Columbia, she landed an early seat at OpenAI during the run-up to ChatGPT, and had a front-row view of a venture industry that, in her telling, was still funding NFTs while the actual technological shifts were happening in labs nobody was paying attention to.Jenny is now a Partner at Leonis Capital, a research-driven fund she co-founded straight out of OpenAI to fix the problem she kept running into: most VCs don’t fully understand the technology they’re investing in.In this episode of Going Solo, we discussed:* The best AI products always arrive ahead of the model. Lovable started as GPT-Engineer years before Sonnet 3.5 made vibe coding actually work. Jenny only backs founders who can see 12-18 months down the model capability curve, because the ones building for today’s models are usually building something a bigger lab commoditises within a year.* Leonis’ algorithm is better than they are at picking winners. Leonis backtested its internal founder-scoring system against 20,000 companies, and it beat the partners. Everything the fund knows (CRM data, memos, judgment) lives in a single GitHub repo so an AI agent can query it instead of asking a human where a five-year-old file went.* “Founder first” has become “founder only”. Jenny argues that plenty of investors now use “I just back great people” as cover for not understanding anything else about a market. Her question: how do you even prove you’re a good judge of character?* Most “neo labs” are the wrong shape for venture money entirely. Leonis mapped 119 of them across capital needs and speed to revenue. Only 8% fit the traditional venture model; 60% need OpenAI-sized capital without OpenAI-speed monetisation, which is a much better fit for sovereign wealth than a ten-year VC fund. We also got into why open-source models are closing the gap on closed-source ones faster than anyone expected, why she thinks Anthropic’s current lead is more fragile than the market is pricing in, and her view that the next ten years of AI is less about assistance and more about software making decisions on our behalf.Hope you enjoy it.DG This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit blog.joinodin.com

This is part two of my conversation with Saul Klein, co-founder of Phoenix Court Group. Saul talked about his investment in Wise and why he’d back founder Taavet Hinrikus to chop lettuce. The story started eight years before Phoenix Court actually signed a term sheet. Saul had worked with Taavet at Skype, so by the time Wise turned up at a Seed Camp demo day, his conviction had nothing to do with the pitch.Phoenix Court is turning founder evaluation from an art into a science. They have spent eleven years building an internal system called Nazare, which collects sentiment across ten dimensions from everyone who meets a founder, and is starting to use that history to flag when a team might be about to repeat a mistake it has made before.There’s more that makes Phoenix Court unique. For Saul, fund size should be treated as a discipline rather than a fundraising target. Local Globe stays at roughly £150 to £170 million, Latitude around £200 million, and Solar around £250 million, sizes Saul says the firm has committed to for its next three fund cycles regardless of what LPs might prefer. Basecamp, the vehicle Phoenix Court uses to back emerging managers off its own balance sheet, charges no fee at all.Saul’s advice to anyone starting out as a solo GP? Get started. For the first couple of funds, he says, the only thing LPs can really diligence is whether you get into good companies and whether founders rate you, and neither requires a large cheque.We finished on where Saul thinks this all goes over the next twenty years, including his hope for a version of the Silicon Valley model that isn’t winner takes all, and what it would take to build many new Palo Altos.Part one covers his path from LoveFilm and Skype into venture, and the recent decision to restructure Phoenix Court as a company with a foundation as its largest shareholder. Worth listening to that part first if you haven’t already.Hope you enjoy it.PR This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit blog.joinodin.com

A few weeks ago, the legendary investor and entrepreneur Saul Klein joined me in the Odin offices. Saul is co-founder of Phoenix Court Group, which runs Local Globe, Latitude and Solar, and has backed the likes of Wise, Monzo, Mistral and Figma. It was a long conversation, which we’re publishing in two parts.In part one, we dive into Saul’s unusual journey into venture: from co-founding DVD-by-post business Video Island, to joining Skype in its early years, before investing at Index Ventures and eventually starting what became Phoenix Court.Knowing the right time to step aside and let others lead. Saul left LoveFilm to join Skype not long after the merger that created it, and has since handed several Local Globe-backed businesses over to operators he considers better suited to running them than he is. His logic each time was the same: if there are people who can do the job better, get out of their way.Phoenix Court recently restructured to become a limited company rather than a partnership, a move only a very small number of the 20,000 active VC firms globally have made. Being a limited company, rather than an LLP, lets the firm keep profits on its balance sheet instead of distributing them to partners every year. Saul argues this enables longer-term decision making. The firm’s largest shareholder is now a foundation that grants most of its funding within a mile of the office.A common theme in our Going Solo episodes is how seed investing is one of the most commoditised businesses there is. With around 20,000 active VCs offering essentially the same product, cash, Saul compares the category to a street of identical hairdressers. The real question for any fund is why a founder should choose to work with you over any other shop on the same road. Saul cites research showing that just 2.4 percent of the stocks traded globally since 1990 accounted for effectively all the value created in public markets. His own numbers for seed are similarly narrow: a 1.3 percent chance that any given investment reaches $100 million in revenue, and a 1.8 percent chance it becomes a unicorn. By his count, only 21 of roughly 20,000 active funds worldwide have picked more than ten of those companies at seed.He borrows a line from Benchmark’s Andy Rachleff: back surfers, not waves. A big enough trend, AI, cloud, whatever comes next, will always attract thousands of investors. The actual job is picking founders who will still be riding waves twenty years later, which is why Phoenix Court’s whole model is built around backing people as early as possible.Part two will come out next Sunday, and covers the origin story behind backing Wise, how Phoenix Court has spent eleven years building an AI system for evaluating founders, and Saul’s advice for anyone starting out as a solo GP. If you can’t wait to watch it, you can check out the whole conversation on YouTube.You can also check out Saul’s Substack, here.Hope you enjoy it.PR This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit blog.joinodin.com