Software Engineer: Agentic Software Development

The job

You build our agentic engineering org: the systems and ways of working that let a small team direct agents to build, review, and ship reliable software. You own how human judgment turns into clear tasks, how work runs in parallel, how results are verified, and how what we learn from each change improves the next one.

Problems you’ll work on

These are examples, not lanes. You will move between them, while taking real ownership of the ones that match your interests.

• Agents as developers: A growing share of our code is written by agents and reviewed by people. What do a repo, a CI system, and a review process look like when the author never gets tired and is sometimes confidently wrong? How do you make a green build mean something?

• Human judgment at agent scale: A person can direct far more work than they can inspect line by line. How do you capture intent, give each agent clear boundaries, and spend human attention on the decisions that need it?

• Fast feedback at monorepo scale: One TypeScript monorepo holds a desktop app, a daemon, cloud services, and dozens of plugins. How do you keep builds, tests, and type checks fast enough that nobody waits on them, including agents running in parallel worktrees, without letting coverage rot?

• An engineering org that improves: Every shipped change, failed build, and review correction teaches us something. How do you turn that evidence into better tools, instructions, and workflows, then check whether the next task benefits?

What you’ll do

• Design and build our agentic development workflow, from scoping and delegation through implementation, review, and release.

• Own the build: CI, caching, test infrastructure, and release tooling. Make it fast, and make its signal trustworthy.

• Build the tooling that lets agents work well in our codebase: parallel worktrees, review gates, evals, and the instructions and skills that tell an agent how we build.

• Turn failures and corrections into reusable skills, instructions, and tooling, so the next piece of work goes better.

• Measure the developer loop: quality, throughput, cost, and the human attention each change needs. Find where the time goes, and remove it.

Who you are

• You’ve owned developer tooling that other engineers depended on every day: a build system, an SDK, a CLI, or CI at scale.

• You think about how a whole engineering team works, from how tasks are scoped to how changes are verified and released, and you can turn that into systems other people use.

• You build with coding agents every day, and you have strong opinions about what makes a codebase easy or hard for them to work in.

• Deep TypeScript and Node. You’re at home in monorepo tooling (pnpm, Turborepo, or similar), and you know how to make a slow build fast.

• You’d rather delete a step than document it.

Bonus

• You’ve introduced agentic development practices across an engineering team.

• You’ve built developer tools designed for agents to use, such as MCP servers or agent-friendly CLIs.

About Yaklabs

Yaklabs is building a desktop AI application that turns knowledge workers into AI power users, without docs, API keys, or knowing what to ask for. One surface for agentic AI work: proactive, personalized, progressively powerful. The gap between AI power users and everyone else isn’t talent, it’s setup. We close it.

The team: Alex Goddijn spent four years at Palantir, then founded an agent harness company that was acquired by Poolside AI. Seb Goddijn led internal AI at Ramp. We’re building Bonsai so that kind of leverage isn’t limited to companies with their own internal AI team.

We have funding. Product and design work in person in New York City, and some engineering roles are remote. Small team, high intensity, no passengers.

Who we look for

• Technical builders with a domain edge. Everyone here is technically minded and can build and ship. You bring a distinctive strength in a domain: design, systems, product, infrastructure, or another area, and use it to take a problem from first principles to working software.

• First principles thinking. We’re building in a category that’s being defined in real time. Best practices don’t exist yet. You reason from the problem, not from what everyone else is doing.

• High agency and ownership. You don’t wait to be told what to do. You see what needs to happen, you go make it happen, and you own the outcome. At a team this size, there’s nobody to delegate to.

• Deep curiosity about AI tooling. Not just using ChatGPT or Copilot: you’re experimenting, writing your own tools, trying new agent frameworks the week they drop, forming opinions about what works and what doesn’t. The people building this product should be the most demanding users of the category. If you’ve shipped PRs to open-source agentic harness projects, tell us.

• You build compounding leverage. In the era of building with agents, one person doesn’t just ship one person’s work, they build the systems that build the product. Every person is amplified, and with that amplification comes the responsibility to build things that compound: skills, automations, agent workflows that make the next thing faster. This is a new way of working, and you’re excited to develop the skillset it demands.

The deal

• Early team. $215k to $325k base salary for US-based hires, plus meaningful equity.

• In person in New York City for product and design roles. Some engineering roles are remote, and each posting lists its location.

• You’ll shape the product, the culture, and the company. This is a founding-team role.

YakLabs is an equal opportunity employer. We consider every application without regard to race, color, religion, sex, gender identity, sexual orientation, national origin, age, disability, veteran status, or any other protected characteristic. If you need an accommodation at any point in the process, email [email protected] and we will work it out with you. A person, not a machine, reads every application.