Member of Technical Staff, Deeptune

About the role

This role operates at the intersection of backend infrastructure and agentic AI, where scalable systems meet real-world task execution. You will partner with researchers and AI companies to shape the next generation of agentic models. The position demands high ownership, hands-on technical leadership, and consistent in-person presence at the New York City office.

Software engineers turn product ideas into working code. Engineers work in small teams, review each other's work, and ship in small batches. Most teams follow agile practices such as sprints and daily standups. Engineers also write tests, fix bugs, and improve performance. The field values clear communication as much as technical skill. Engineers spend part of every week on planning, code review, and debugging, not just writing new code. The ability to explain a technical decision in plain words separates strong engineers from the rest.

Key facts

What you'll do

Python, Go, and TypeScript build scalable backend and infrastructure systems that power agentic AI training environments.

Ambiguous, fast-moving environments are navigated using sound technical judgment to set direction and drive outcomes while staying hands-on.

Engineering teams are coached, and execution is driven through strong ownership and pragmatic decision-making.

Five days a week in-person work at One World Trade sustains deep collaboration and immersion in the Deeptune lab.

Requirements

A bachelor's degree is required.

You must have 10+ years of engineering experience, with 2-3+ years focused on post-training, evaluation metrics, and reward modeling for LLMs.

You must demonstrate a strong background in building scalable systems that support high-throughput AI training workflows.

You must possess deep comfort with core ML and LLM concepts to partner effectively with researchers rather than only consume their outputs.

You must thrive in ambiguous, fast-moving environments and apply sound technical judgment to prioritize and execute.

You must have a proven ability to coach engineers and drive execution across cross-functional initiatives.

You must show a strong ownership mindset with pragmatic decision-making aligned with product and research goals.

Practical notes

Work is performed in-person five days a week at the One World Trade office in New York City.

A $15,000 relocation bonus is provided for moves to the New York metropolitan area, and a $10,000 housing bonus applies when living within 0.5 miles of the office.

A $1,500 monthly stipend covers meals, while Free Equinox membership, $200 monthly laundry reimbursement, and $200 monthly personal wellness reimbursement are included.

Health, Dental, and Vision insurance are provided together with 401(k) contributions and company match.

Typical interview steps

Hiring for engineering roles usually starts with a recruiter screen, followed by one or two technical rounds. Candidates often solve a coding problem, discuss past projects, and answer system design questions. Some loops include a take-home task. Final rounds typically cover team fit and give candidates a chance to ask questions. Interviewers look for how you break down an unfamiliar problem, not just whether you reach the answer. Practicing a few problems aloud and reviewing your own past projects are the best preparation.

Good to know

The role focuses on backend infrastructure, scalable systems, and reinforcement learning environments for AI agents.

Tools commonly used include Python, Go, TypeScript, and frameworks for reward modeling and evaluation.

Success depends on craft mastery, ownership, and comfort with ambiguity in a fast-paced setting.

The team values diverse perspectives and uncommon ideas while building technology that impacts how expertise powers AI advancement.

About the company

Mercor is defining the future of work. We partner with leading AI labs and enterprises to provide the human intelligence essential to AI development. Our vast talent network trains frontier AI models in the same way teachers teach students: by sharing knowledge, experience, and context that can't be captured in code alone.