GPU Programming Software Engineer
- Pay
- $60 – $95 / Hour
- Open to
- Worldwide
- Apply
We earn a commission if you sign up through the links on this page. It costs you nothing and does not affect which jobs we list. How this works.
- Skills
- cuda
- webgpu
- glsl
- c++
- gpu programming
- performance optimization
What you'll do
You write GPU programming tasks that teach language models to reason about performance and hardware architecture. In practice that means implementing and optimising GPU code in CUDA, WebGPU or GLSL, profiling kernels and shaders to squeeze out performance, and writing the C++ host-side logic that drives them.
The ad is short on process detail, but the output is clearly task material rather than production software: problems with a correct, fast solution that a model can be tested against. The interesting tasks in this space are usually the ones where the naive version works and the fast version requires understanding memory access patterns, occupancy or synchronisation.
Who fits
- Advanced experience in a GPU language that runs on NVIDIA hardware
- A background in graphics programming, ML acceleration, scientific computing, HPC, or another GPU-heavy field
- Proficiency in C++
That is the whole published list. No degree, years of experience or English level is stated, though the work will be in English.
What it pays
$60–95/hour as published, but compensation is output-based: you are paid per task that meets the project specifications, and a minimum number of tasks per week applies. The per-task fee and the weekly minimum are not given.
For comparison, micro1 pays its general ML Engineer listing $100–150/hour. GPU specialists can command more in industry than this band suggests, so this makes most sense as flexible side work.
Worth knowing
Good:
- Broad entry: CUDA, WebGPU or GLSL, and several domain backgrounds count
- Graphics programmers, not only ML people, are explicitly welcome
- 25 openings, flagged as high demand
- Fast start: micro1 says roles fill within 48 hours and first tasks begin 24–48 hours after onboarding
Less good:
- Per-task pay with an unpublished weekly minimum
- Rate is modest for specialist GPU skills
- Very little detail on what a task contains or how it is reviewed
- Contractor engagement: no benefits, your own tax to manage
About this listing
Posted by micro1 on its own jobs portal and read on 24 September 2026. The ad's own role title is "GPU Programmer Software Engineer". Location is Remote with no country list. See micro1.
More roles at micro1
See all 380Similar roles at other platforms
- Lean Engineer, Formal Mathematics (Lean 4, Mathlib, Theorem Proving)Mercor · $90 – $110 / Hour · 10d ago
- Software & Firmware Engineering ExpertsMercor · $100 – $120 / Hour · 14d ago
- Senior Software Engineer: AI Evaluation & BenchmarksAlignerr · $80 – $100 / Hour · 1mo ago
- Technical Task AuditorInvisible Technologies · $60 / Hour · 28d ago
Guides about micro1
See all 81- micro1 Computer Vision Engineer Interview: Transfer Learning, Preprocessing and DetectionInterview prep
- micro1 Data Engineer Interview: ETL Reliability, Data Quality and Warehouse CostInterview prep
- micro1 AI Engineer Interview: Feature Engineering, Overfitting and Deploying ModelsInterview prep
- micro1 DevOps Engineer Interview: Continuous Integration, IaC and OrchestrationInterview prep
Browse similar roles
Not the right fit?
See every open role, or get new ones on Telegram or Discord as they are added.