Senior Research Engineer - AI Coding Tools
Build AI coding agents, fine-tune code LLMs, and extend the ComputeEval benchmark for CUDA developers.
Originally posted 30 October 2026 by the employer. Relisted by the employer on 31 October 2026.
About the role
NVIDIA's AI Developer Tools organization is seeking a Senior Research Engineer to build AI coding agents, models, datasets, and evaluations for CUDA developers. The team prototypes and ships novel coding agents, fine-tunes code LLMs, publishes benchmarks like ComputeEval, and integrates core developer tools such as Nsight Compute and Nsight Systems.
What you'll do
- Build and improve novel coding agents that help developers write, optimize, and maintain CUDA code within developer workflows.
- Design and ship evaluations, including extensions of the public ComputeEval benchmark, for CUDA development.
- Fine-tune and specialize code LLMs, partnering with the Nemotron team on training datasets and evaluations.
- Develop Agent Skills, MCP servers, and tool-use interfaces that integrate developer tools like Nsight Compute and Nsight Systems with AI agents.
- Generate, curate, and validate synthetic training and evaluation data for CUDA programming.
- Deliver new knowledge to LLMs through RAG and skill-based systems, and collaborate with product teams to turn research prototypes into shipping features.
What you'll need
- B.S. in Computer Science or related technical field, or equivalent experience.
- 12+ years of industry experience in applied AI/ML, with recent work on coding agents, code LLMs, or AI developer tools.
- Strong proficiency in Python and sound software engineering practices.
- Hands-on experience fine-tuning or evaluating LLMs.
- Fluency with LLM-powered agent systems, including context management, prompt caching, tool-use design, MCP, and Agent Skills.
- Experience designing or contributing to rigorous evaluations for code generation or agentic systems, with a track record of shipping software to real users.
Nice to have
- Public contributions in the AI-for-code space, such as open-source agents, benchmarks, papers, or technical blog posts.
- Experience building coding agents or code LLMs used by real users daily.
- Familiarity with CUDA, GPU programming, NVIDIA profiling tools (Nsight Compute, Nsight Systems), or libraries (cuDNN, cuBLAS, Thrust, CUB).
- Experience with synthetic data generation and quality validation for code.
- Track record of zero-to-one product work.
Skills: AI coding agents, LLMs, CUDA developers, ComputeEval, developer tools
More semiconductor jobs: NVIDIA jobs · Software engineer jobs · California jobs
This role has been open 182 days — well above the 67-day median for Software Engineering Semiconductor roles.
Software Engineering Semiconductor · AI ML Hardware
|
Open roles in category
148
|
Median days open
67 d
|
Category median salary (all employers)
$188k
|
See the full market breakdown ▾Category comparison, skills in demand, and who else is hiring
| Metric | NVIDIA | All employers we track in this specialty (148 roles · 43 employers) |
|---|---|---|
| Open roles in this specialty | 34 | 148 |
| Open roles in the wider Software, Firmware & Systems family | 1117 | 6358 · 144 employers |
| Median days open | 229 d | 67 d (+162 d vs this employer) |
| Median salary (USD postings) | — | $188k |
Skills observed across this category: AI coding agents, LLMs, CUDA developers, ComputeEval, developer tools
Who's hiring in this category
- NVIDIA (this employer) · 25 open roles · median 229 d
- Micron Technology · 23 open roles · median 56 d
- Qualcomm · 18 open roles · median 205 d
- Infineon Technologies · 12 open roles · median 50 d
- AMD · 8 open roles · median 67 d
- Arm Holdings · 4 open roles · median 3 d
How we counted: 148 open Software Engineering Semiconductor (AI ML Hardware) roles from 43 employers tracked in the SemiconductorJobs index, counted 31 Oct 2026. Specialty figures count only roles carrying this exact specialty label, so an employer's related work in neighbouring specialties is not included there — it is counted in the wider Software, Firmware & Systems family row. Figures refresh nightly.