Founding Director of Engineering - AI-Native Venture Studio

In one line: We build many products at once with small teams of humans + AI coding agents. You run that engine — you lead the product managers, orchestrate the agent fleets, set the technical bar, and still ship code yourself. Player-coach, not a hands-off manager.

Why this is one of the best seats in AI right now

We're an AI-native venture studio. We stand up new software products — our own and for clients — fast, on top of an in-house agent platform (per-chat sandboxes that scaffold, build, test, deploy and database whole apps; subagents; MCP/skills). Our operating model is humans and AI coding agents on one board: work gets assigned to agents like teammates, and a small team ships what used to take a big one. You'd be the person who makes that machine run across the portfolio.

What you get here that you won't get most places:

• Real scope, real ownership. You own delivery across multiple products, lead the PMs/builders who run them, and are the technical truth-source the team ladders up to. This is a build-the-org seat, not a maintain-the-org seat.
• No budget cap for tokens, no babysitting. Best-in-class tooling, frontier models, every AI coding harness and agent framework you want — and nobody counting your tokens. If you've been throttled by a $20 plan while doing real work, that ends here.
• A brilliant CTO to build alongside. You report to a CPO/CTO who is genuinely excellent and hands-on; you'll get sharper fast, and you'll have the air cover to move.
• Frontier work, not maintenance. Agentic systems, net-new products, an internal platform that builds other products. You'll work with the newest models and harnesses the week they drop.
• Top-of-market pay, paid properly and on time. We pay our best people what they're worth and we don't play games with it. (See comp below.)
• No politics, no process theater. We're direct with each other — genuine, but not soft. Speed, output, and quality beat ceremony.

What you'll actually do

• Lead the product managers / builder-leads. Each owns a product or venture; you set direction, unblock, review, and raise their tempo and quality. You hire and grow this team.
• Orchestrate humans + AI agents. Break product bets into work for people and for agent fleets, decide what's an agent job vs. deterministic code, and get correct, shipped output back.
• Stay hands-on. You still architect and write code — the hardest plumbing, the risky spikes, the reference implementations. You out-build when you need to.
• Own the portfolio's delivery. Multiple products shipping in parallel, fast, without it turning to mush. You make the build-vs-buy-vs-agent calls and the architecture decisions.
• Set the standards that let a small team ship safely at speed — CI, review, evals, security and logging conventions, the guardrails that keep quality high while agents and people ship weekly.
• Reason commercially. Figure out why a product or funnel isn't working from first principles, and fix the real cause — you think about the business and the user, not just the code.

Requirements

Who you are

• You're a player-coach. You lead people and you still ship. A leadership title where you can no longer build is not what this is — and not what you want.
• You're an agentic-engineering native. You don't just use AI — you build with it at the systems level: agents, memory, tools, orchestration, evals, failure modes. You use Claude Code / Codex / Cursor and agent harnesses daily and push them to their limits.
• You get leverage through others. You've led engineers and/or product managers and gotten a team (or a fleet of agents) to produce far more than you could alone — without becoming a bottleneck or a status-chaser.
• You've shipped, fast, more than once. Ideally you've founded or led 0→1 products. You've turned around in a weekend what others quote in months, and you set that pace for a team.
• You see the whole board. Technical, but you reason about the business and the user. You can take a fulfillment gap and see the revenue model, not just the bug.
• You're right, and so is your team's work. You install the standards that mean shipped work is tested and works — nobody has to check whether it's 50/50 wrong.
• You're high-agency. Drop you a one-line product bet and you make sensible calls, document assumptions, and ship — and you make your team do the same. You don't wait to be managed.
• You can take a punch and throw one. We're blunt; we'll attack an approach hard because that's how the work gets good. You engage, defend your thinking with substance, and bounce back — and you set that healthy-blunt culture for your team.
• You live in the frontier. Your feed is AI all day. You can name what you started using in the last two weeks and why.

Signals we screen for (show us the strongest — you don't need all)

• You've led an engineering and/or product team — managed engineers and/or product managers, or run eng as a technical founder/CTO/Head/Director of Engineering — ideally at a fast, AI-native company.
• Hands-on agentic depth — you build with LLMs and agent frameworks (Claude Code, Codex, Cursor, MCP, LangGraph/CrewAI, RAG, evals, n8n AI automations), not just prompt a chatbot.
• A portfolio of shipped products — things you founded, led, or personally built and drove. This matters more than your résumé or degree (some of our best people have non-CS backgrounds).
• Commercial fluency — startup / agency / B2B-SaaS / GTM context; you reason in revenue and interrogate ICP and offer.
• Range — you can go from architecture and code review to a live product decision to a hard conversation with a client, in the same day.
• Bonus: experience orchestrating AI agents across multiple products, martech/sales-tech depth, or a venture-studio / forward-deployed background.

What this role is NOT

• Not a hands-off manager. If you haven't shipped code in two years and don't want to, this isn't it.
• Not a pure IC. If you'd rather build alone than get leverage through a team and agents, this isn't it either.
• Not a process/ceremony role. We want guardrails, not gates. No sprint theater.
• Not a place that needs all-nighters. We want you thinking at your best. Healthy obsession, not grind-cope.

Benefits

Compensation

For the genuine article this is a top-of-market seat: roughly $15,000–$40,000 / month USD (~$180K–$480K annualized), contractor, paid monthly and on time. Where you land in that range is about the tier you hit — a true player-coach who leads PMs and orchestrates agents across products sits at the top — not where you live. We pay for the talent tier, not the postal code, and there's real upside on what you own. International candidates: a strong USD monthly rate goes a long way, and we don't lowball the real ones.

Logistics

• Type & framing: We start as a paid engagement — the best leaders we hire often run their own thing and we respect that — and it grows into whatever's right for both of us: bigger scope, a real stake. This is a peer seat, not a probation.
• Location: Remote, anywhere with enough daily overlap to actually lead the team. We work US/Europe-centered hours; Americas or EMEA overlap is easiest, but we'll flex for the right person.
• Exclusivity: Flexible to start; the strongest engagements grow into primary over time.

How to apply

No long form, no unpaid take-home busywork. We want to see what you've built and how you lead — a tight Loom(keep the recording under ~15 min; think as long as you like):

1. Show us something real you led or built — a product you founded/led, a team you built, an agent system you shipped. Walk us through the hardest problem, the calls you made, and where it still breaks. Live links we can poke at are ideal.
2. A studio teardown: "You're handed 3 new product bets, 2 product managers, and a fleet of AI coding agents, and asked to ship all three to a working v1 in 30 days. How do you set it up?" Hit team + agent orchestration, the standards/guardrails you'd install, your architecture and build-vs-buy-vs-agent calls, what you'd cut, and your honest weakest point. (Some of this is deliberately underspecified — make calls and state your assumptions.)
3. One line: what's something in AI/agents you started using in the last 2–3 weeks, and why?

Send your link(s) to us. If it's a fit, expect one live session: we'll be direct, we'll change a requirement on you mid-conversation, we'll have you whiteboard or build something small live, and we'll push hard on your thinking. If a recorded video isn't your style (some who run their own thing prefer not to), tell the recruiter — we'll do a live equivalent or jump to a short paid build.