Remote Staff Scientist, Coding Agent Evaluation & Benchmarks

Remotebridge; United States; Remote; USD 200,000 - 260,000; Full time

Benefits offered by this job

Equity compensation

Performance-based bonuses

Health-insurance premium reimbursement

401(k) company match

Remote-first workforce

Job summary

micro1 is seeking a Member of Technical Staff to advance evaluation and development of frontier coding agents. You will design benchmarks, methodologies, and data systems shaping how next-generation coding models are measured and improved.

Split across AI research, software engineering, and model evaluation, you will drive end-to-end research, build datasets and protocols, and work with cross-functional teams in a remote-first environment.

Qualifications

• Proven software engineering background with strong Python/C++ skills.
• Experience in ML research, model evaluation, or related disciplines.
• Ability to design and review benchmarks, tasks, and evaluation protocols.
• Strong analytical skills to interpret model behavior and failures.
• Excellent written and verbal communication for diverse audiences.

Responsibilities

• Design and own evaluation frameworks for coding agents, including benchmarks and rubrics.
• Lead end-to-end research initiatives to measure and improve coding model performance.
• Develop datasets, golden examples, and evaluation protocols for reliable assessments.
• Analyze model behavior and failure modes to identify weaknesses and improvements.
• Build tooling and infrastructure for large-scale experiments and data pipelines.
• Establish best practices for coding-agent assessment, ensuring rigor and reproducibility.
• Collaborate with researchers, engineers, and AI teams to design experiments and communicate results.

Skills

Python

C++

Software Eng

ML Research

Evaluation Methods

Job description

micro1 is seeking a Member of Technical Staff to advance evaluation and development of frontier coding agents. You will design benchmarks, methodologies, and data systems shaping how next-generation coding models are measured and improved.

Split across AI research, software engineering, and model evaluation, you will drive end-to-end research, build datasets and protocols, and work with cross-functional teams in a remote-first environment.