Salary range: $300k - $350k | Equity: 0.4% - 0.6% | In-Person: NYC

About Datalab

Datalab builds the models that read documents reliably at scale. The world's most important information is trapped in PDFs, scans, and files that software can't actually parse, and getting it out correctly matters. From frontier AI labs processing training data to Fortune 500s like Siemens extracting decades of engineering records, Datalab is where businesses turn to when extraction has to be right.

We’re at an 8-figure run rate with a team of 7. Anthropic is a customer. And we have hundreds more across FAANG, frontier AI labs, healthcare, finance, government, and legal. Our models, Chandra, Surya, Marker, and Lift, have 70,000+ GitHub stars and broad developer mindshare. We're backed by founding members of OpenAI, FAIR, and Hugging Face.

Do the math on the first two facts. Every person here carries more revenue than most startups generate in total. This is the kind of leverage a role at Datalab means.

Role Overview

We're looking for an engineering lead to guide our team while staying hands-on in the code. You'll set the technical direction and standards for how we build the interfaces, tools, and infrastructure behind our OCR, extraction, and document-understanding systems. This includes everything from optimizing agent loops and interfaces to helping to speed up inference.

This is a player-coach role. You'll manage and grow a team of three engineers, own engineering delivery and quality, and spend a large share of your time writing code - focused on architecture, infrastructure, and the hard problems rather than routine feature work. You’ll partner closely with the research team to define the handoff between experimentation and production.

As a small and fast-moving team, roles are fluid and ownership is high. You'll work directly with the founder to set priorities, ship features, and make our technology accessible to a global community of builders.

Day to day, you will:

• Manage and grow a team of three engineers - 1:1s, prioritization, feedback, and hiring as we scale.
• Own engineering delivery, quality, and technical standards across code, testing, infrastructure, and deployment.
• Stay hands-on (~40–50% of your time), contributing architecture, infrastructure, and code review across our open-source repos, API, and internal tooling.
• Contribute to the research-to-production interface: set standards for turning research prototypes into reliable systems.
• Partner with the research team to align on what's ready to productionize and how.
• Help set engineering priorities and the roadmap alongside the founder.

Ideal Candidate

You've led a small engineering team while also shipping features. You're energized by both growing engineers and solving hard technical problems yourself, and you know how to set standards that people actually adopt. You operate with autonomy and can navigate the ambiguity of an early-stage team where the research and engineering boundary is still being drawn.

We're eager to work with someone who has:

• 7+ years of engineering experience, including 2+ years leading or mentoring engineers.
• A track record shipping and maintaining production systems serving high-volume traffic.
• Strong fullstack or backend depth, with the judgment to set architecture and technical standards.
• Experience building with LLMs and agentic workflows.
• Experience building in early-stage startup environments.

Bonus points:

• Experience partnering closely with research or ML teams, or otherwise bridging research and production.
• Familiarity with document processing, computer vision, or OCR systems.
• Maintained or contributed to open-source projects with active communities.

Interview process

1. 30-minute video call to evaluate fit
2. 1.5-hour in person meeting to build an architecture together
3. Async code sharing + review phase (not a take-home project, you just share some code samples)
4. Culture fit interviews with the team

At this stage of the company, every interview is somewhat custom, so these phases may be rearranged slightly.