Eniac Ventures

From our first fund in 2010 at $1.6m to announcing $220m today, we remain steadfast in our mission to serve founders at the earliest stages. We are grateful to our founders, LPs, co-investors and the community for letting us continue doing what we love. techcrunch.com/2024/04/24/see…

The latest Carta data on $70B+ invested into Seed through Series C startups in 2025 highlights something many of us in NYC see every day. NYC is now: • #1 in Fintech • #1 in Consumer • #2 overall startup ecosystem • Top 2 in AI, SaaS, Biotech, and Healthtech But the most interesting takeaway isn’t the rankings, it’s the supply and demand imbalance. Founder demand in NYC still significantly exceeds early-stage VC supply. Which makes NYC one of the largest structural arbitrage opportunities in venture today. 👇🏾👇🏾👇🏾

What do you get when you put the best founders you can find in one room? At least $192 billion in value created (!!) and some really good pasta. Last night, @dhaber and I cohosted an @a16z and @EniacVC "Human Unicorn Dinner' at the acclaimed Torrisi in NYC — an intimate gathering of founders and operators who have built billion-dollar companies, navigated acquisitions, and rebuilt from scratch. In the room: Operators who scaled Axiom Space, ZocDoc, Maven, Databricks, Notion, Stripe, Reddit, Muse, Bond, Bluecore, mParticle, Moloco, Metaplane, Wealthsimple, Slice. Founders now building the next wave — Upscale, PlanZ, Octogen, Crossterra, Trove, Sitelevel, and more. But the number that hit me hardest wasn't the valuation figure. It was the quality of the conversation. We went deep on AI — the real stuff, not the headlines: → Has AI lowered the bar to build a unicorn, or raised it? → What does it mean to be a 'technical' founder when AI can write the code? → How do we take care of humans after the AI job disruption? We also talked about when to sell and when to keep going. What it really means to 'win.' And laughed about the failures nobody writes about. Here's what I keep coming back to: building community with founders is critical, especially at the highest levels where it can often be the loneliest. That's why rooms like this matter so much. No pitches. No performance. Just people who've actually done it, being real with each other over (really) good food. Thank you to every founder who showed up and helped build this community of founders operating at the highest levels, right in our backyard in New York City. Can't wait until the next one! 🗽🦄🍝

Many LPs still don’t understand one of the biggest advantages of seed-focused funds. It’s much easier for us to sell secondaries compared to large multistage funds. We’ve returned entire funds to LPs solely from selling secondaries in rounds where the larger funds who followed us couldn’t sell for signaling reasons. When a multistage fund sells, the market notices. When a seed fund takes chips off the table alongside the founders, it’s expected. Real returns to LPs, not just markups on paper.

At @EniacVC we’re investing in companies built for a world of abundant intelligence. At the agentic layer, that means leaning into regulated industries like healthcare, government/defense, and financial services, where data friction is the moat, not the obstacle. The companies that will win here aren’t just building smart products. They’re willing to take on liability, to stand behind the output, own the outcome, and put skin in the game alongside their customers. At the tooling layer, it means things like security, privacy, and compliance. The model companies can’t be the ones solving this. You don’t grade your own homework. These need to be independent layers, and the need only grows as agents move deeper into mission critical environments. In the age of abundant intelligence, the hardest areas to build in will likely be where the most durable companies emerge.

Fantastic @nytimes feature (s/o @eringriffith!) this week on @eniacvc portco @evertune_ai - helping brands navigate marketing in an increasingly AI native world: "The change has also meant new opportunities for entrepreneurs like Mr. Stempeck, who co-founded Evertune in April 2024. Since then, the company has raised $20 million and has more than 200 clients.... Chatbots are insatiable for detailed information, particularly in areas where there is a void, Mr. Stempeck said. So businesses that want to influence what the programs spit out can flood the zone with minutiae. One luxury fashion brand went from publishing five pieces of content a month to around 100, he said, while automakers made sure to publish their entire owner manuals."

Today marks the 80th anniversary of ENIAC, the namesake to our firm and one of the most important milestones in computing history. The co-founders of @EniacVC including myself met at the University of Pennsylvania in 1995, where ENIAC was built and unveiled in 1946. At the time, it was about 1,000x faster than prior mechanical computing methods and could perform roughly 5,000 calculations per second. What once took days could suddenly be done in minutes. Some wild perspective on that machine: • ~18,000 vacuum tubes • ~30 tons of hardware • ~1,800 square feet of space • ~150 kW of power consumption • Millions of hand-soldered joints And yet it started the digital age. Fast forward to today and a single NVIDIA GPU is estimated to be 25 billion times faster than ENIAC. The slope of computing progress since 1946 is almost hard to comprehend. What inspires me most is not just the speed gains, but the mindset. ENIAC’s creators were pushing the limits of what seemed possible, building for a future they could barely imagine. As investors focused on the next generation of builders, it feels fitting that our firm carries that name. A reminder that today’s breakthroughs can become tomorrow’s infrastructure. Here’s to the pioneers who started it and all of you founders pushing the insane frontier of AI today. 🚀

At @EniacVC, we do one thing: lead seed rounds and then help founders find product-market fit. Everything we build as a firm is optimized around those outcomes. That's why I'm really excited to share that @_dorianruth has joined us as our Head of Platform. Dorian brings a great mix of venture experience and GTM execution, exactly the combination that matters at the seed stage. He'll be working directly with our portfolio founders on business development, fundraising prep, customer intros, and the kind of strategic support that actually moves the needle early on. If you've built a company, you know how much of the hard stuff happens between the seed and the Series A. Having someone like Dorian fully dedicated to helping founders navigate that stretch is a big deal for us. Welcome to the team, Dorian!

The highest form of NPS in venture is when a founder you’ve backed chooses to work with you again. Proud to back Chieh again. We were fortunate to seed him at Boxed and excited to support him now as Pelgo, an AI-powered career transition agent helping people reskill and move into AI-first roles

Proud to have been part of this 🙏🇺🇸

While most VC firms are hiring “Heads of AI” to build internal tooling, at @EniacVC everyone on our team builds. Nobody knows your pain points better than you do. And if you’re investing in AI, you need to feel what’s changing, not just read about it.

Some founders don’t slow down, they reload. I’ll never forget that conversation with @jimpayne a few years ago. He knew he had to get back in the arena to seize the largest computing shift in history and build the AI-native platform to monetize it. After building MoPub and MAX, he wasn’t chasing another company, he was chasing a legacy. That conviction is why we backed him. And today, we’re proud to announce the general availability of @cloudx. LFG! 🚀

I think most people, including some really smart ones, are misunderstanding what's going on with Clawdbot/Moltbot. From what I can tell there is no new intelligence in any of these tools. They're using the same models we had a month ago. It's just a very well thought out system for those models to make tool calls - lane-based queues for reliability, semantic browser snapshots instead of screenshots, sandboxed shell execution, and clean abstractions over different model providers. The innovation is infrastructure and UX, not AI. It's given an easy way for most people to give models computer access, so the volume of tool calls has gone up dramatically. The models could always do this - someone just made it frictionless. This isn't some sort of takeoff. It's just very good software engineering.

We’re in the early days of transitioning to an intelligent agentic web. Eventually I believe you’ll have a single multimodal interface for everything you do (potentially two: one for work, one for personal). That primary interface will be a knife fight between Google, Meta, OpenAI, Anthropic, Microsoft with a potential carve out for vertical specific solutions. The rest of the value will be made up of tool/service/subagent providers. If you provide the best service, the assistant should send traffic your way. In some cases the assistants will try to subsume that second layer, so building lasting moats will be key. Companies that figure out how to be indispensable to the assistant layer will capture enormous value.

I’ve crossed over to using my own Claude Code–built tooling far more than the SaaS tools we use at @EniacVC Our SaaS tools are quickly becoming backend datastores for my assistant, not the primary interface How it works: I wrapped a set of APIs into MCP servers that I host locally on my laptop • Gmail: read + send emails, intelligent bulk read and archive to keep inbox clean • GDrive: pulls context from investment memos, strategy docs, etc • Slack: context + alerts • Airtable: portfolio data and returns • Affinity: deal tracking and network context • Circleback: pulls all my meeting transcripts • GCal: read-only for now • More sources coming, mostly external My main user interface is the Claude assistant that has access to all the above I run cron jobs that continuously scan for time-sensitive emails and alert me in Slack when action is required. This calls the Claude API so it can scan using Opus 4.5 Since Claude only currently supports local MCPs, everything runs in a shared Google Drive folder so it automatically syncs across all my machines. There's a JSON config file that cannot live in Drive so you need to use a symlink for that I've been building this for about a week and a half If I can help with your build-out, ask me anything 👇

We knew this was coming but two days ago @OpenAI made it official - they are rolling out ads into ChatGPT. Every tech wave eventually faces the same question: who pays for it? AI is hitting that moment now. Every major computing wave brings its own foundational models, infrastructure, apps, and monetization layers. We saw it with the web, mobile, cloud, and crypto—and now it’s unfolding in AI. As VC subsidies taper, the need for sustainable monetization will only intensify. Consumers won’t want to shoulder the full cost of AI subscriptions, creating an opening for new advertising models and revenue opportunities. At @EniacVC , we’ve been backing these “picks and shovels” for over two decades- including Admob, Tap, Vungle, @attentiveHQ , @evertune_ai, @upscaleaiHQ , @cloudx - and we’re more bullish than ever in the AI era. If you're building in the AI application monetization space, please reach out as we'd love to jam with you!

I keep hearing predictions that VC will “massively consolidate” and a handful of mega funds will eat the industry. That view ignores both history and venture logic. In private equity, a far more mature market, the top five firms manage only ~20–30% of total PE AUM. Hedge funds and real estate show similar long-tail structures even with iconic giants. More importantly, this consolidation narrative contradicts the core venture belief: we invest in startups precisely because small, focused teams can outmaneuver larger, better-resourced incumbents. If you don’t believe in that, you shouldn’t be in venture.

Do you live in SF but like some of these things? • meeting people from other industries • access to culture • food after 9pm • dating We may have a job for you We’re hiring an investor to be based in NYC with strong SF networks. Technical background strongly preferred. Link to apply confidentially in my bio

It took a few days for the gravity of this to really sink in. The largest acquisition ever by the largest market-cap company in the world. A truly historic moment in AI. TWENTY BILLION DOLLARS. But what makes this one meaningful to me isn’t just the number—it’s the people behind it. I’ve known @sundeep since 2008, when we partnered with his and @avarma's first company and they later invested in mine. Since then, we’ve collaborated on and invested in dozens of each other’s ventures (and counting). Our families have grown close. Over all these years, I’ve had the privilege of watching the journey from a front-row seat—the building, the setbacks, the recalibration, and the relentless focus it takes to do something hard the right way. No shortcuts. Just patient, inspiring, determined, sustained effort. Seeing Sunny and @GavinSherry reach this milestone after everything they poured into it, is deeply gratifying. This outcome is earned in every sense of the word. Moments like this are a reminder that behind every headline is a long arc of trust, perseverance, and belief—built day by day, often far from the spotlight. Huge congratulations to the entire @GroqInc team. This is one for the generational books.