Member of Technical Staff, Coding Research

What you'll do

This is a staff job at micro1 itself, not a contract gig, and it sits at the design end of AI evaluation. You own evaluation frameworks for coding agents: benchmark specifications, scoring methods, rubrics and quality standards. You build the datasets, golden examples and protocols that make assessment reliable, dig into model failure modes and turn them into training and evaluation improvements, and build the tooling behind large-scale experimentation, data generation and review.

The output is partly public-facing: technical reports, benchmark studies and client research. If you have been one of the contractors grading coding-agent output, this is the role that decides how that grading works.

Who fits

• A strong software engineering background in Python, C++ or similar
• Three or more years in software engineering, ML, AI research or evaluation
• Experience designing or validating technical assessments, benchmarks or coding tasks
• Familiarity with LLMs, coding agents, RL or model evaluation
• A habit of building tools and automating workflows through systematic experiments
• Clear writing about technical findings, and comfort with ambiguity

Preferred: frontier-AI or coding-agent work, benchmark or dataset design at scale, agentic and post-training familiarity, publications or open-source contributions.

What it pays

The ad's compensation notice gives a national base salary range of $200,000–260,000 USD for this full-time role, plus eligibility for equity and possible performance bonuses. Benefits include up to 100% reimbursement of health-insurance premiums, paid time off and a 401(k) with company match.

The listing's pay display is inconsistent: the job card shows $600,000–1,300,000/year (which may include equity; the ad does not say) and the detail panel shows an hourly figure that cannot be right for a salaried role. We use the base salary written in the compensation notice. Confirm the full package before you rely on the larger figure.

Worth knowing

Good:

• Salaried, with equity and benefits, unlike micro1's contract listings
• Shapes how coding agents are measured across clients
• Remote-first employer

Less good:

• One opening, and a competitive profile
• The pay display conflicts with the written salary range
• The 401(k) and "national pay range" wording suggests US employment; the ad does not say which countries it can hire in
• The ad notes AI tools are used in candidate screening

About this listing

Posted by micro1 on its own jobs portal 11 days before we read it on 24 September 2026. The base salary range, benefits and requirements come from that listing's text. Visa sponsorship is not addressed. See micro1.