Remote | Member of Technical Staff, Coding Research — $600,000–$1,300,000/year
About The Position
This full-time, fully remote role focuses on evaluating and developing frontier coding agents at the intersection of AI research, software engineering, and model evaluation. The role combines benchmark and dataset design, coding-agent evaluation, model failure analysis, large-scale experimentation, research tooling, and collaboration with researchers and engineers. Responsibilities span research methodology and hands-on technical implementation, with priorities and evaluation methods evolving as model capabilities develop. Work must be completed without using confidential or proprietary information belonging to an employer, client, institution, or other third party. The opportunity is available through 24-MAG LLC, which connects experienced professionals with remote consulting opportunities across technical, evaluation, and project-based workstreams.
Requirements
- Strong software-engineering background and expertise in Python, C++, or comparable programming languages.
- At least 3 years of experience in software engineering, machine learning, AI research, evaluation, or a related technical discipline.
- Experience designing, reviewing, or validating technical assessments, benchmarks, coding tasks, or evaluation methodologies.
- Familiarity with large language models, coding agents, reinforcement learning, model evaluation, or related AI systems.
- Ability to build tooling, automate workflows, and improve technical processes through systematic experimentation.
- Strong analytical skills and ability to investigate complex model behaviour and technical failure modes.
- Excellent written and verbal communication skills.
- Ability to operate effectively in fast-moving research environments with significant ambiguity and evolving priorities.
- Strong interest in understanding how data, evaluations, and feedback mechanisms influence model capabilities.
Nice To Haves
- Experience with frontier AI systems, coding agents, or model-evaluation research.
- Experience designing benchmarks or datasets for machine-learning systems at scale.
- Familiarity with agentic workflows, tool use, reinforcement learning, or post-training methodologies.
- Publications, open-source contributions, or demonstrated technical leadership in AI, machine learning, or software engineering.
Responsibilities
- Design and own evaluation frameworks for advanced coding agents.
- Develop benchmark specifications, scoring methodologies, rubrics, and quality standards.
- Establish rigorous methods for measuring coding-model performance across diverse software-engineering tasks.
- Define objective criteria for correctness, reasoning quality, robustness, and task completion.
- Maintain methodological rigour, reproducibility, and consistency across evaluation workflows.
- Develop high-quality datasets, golden examples, and structured evaluation protocols.
- Design technical tasks and data workflows that reliably assess frontier coding systems and support model development and iterative improvement.
- Identify gaps in benchmark or dataset coverage and develop new evaluation categories as needed.
- Analyze coding-agent behaviour to identify systematic weaknesses, failure modes, and performance limitations.
- Investigate incorrect reasoning, implementation errors, tool-use failures, and incomplete task execution.
- Translate findings into actionable recommendations for model training and evaluation.
- Design experiments to test hypotheses about coding-model capabilities and use results to improve datasets, methodologies, and system performance.
- Build tooling and infrastructure for large-scale experimentation, data generation, review workflows, and evaluation pipelines.
- Automate technical processes to improve evaluation efficiency and research velocity.
- Collaborate with researchers, engineers, applied AI teams, and other technical stakeholders.
- Contribute to technical reports, benchmark studies, research documentation, and external-facing research initiatives.
- Communicate complex technical findings to specialist and broader technical audiences.
Stand Out From the Crowd
Upload your resume and get instant feedback on how well it matches this job.
What This Job Offers
Job Type
Full-time
Career Level
Mid Level
Education Level
No Education Listed
