Loading the job postings and counting them…
Loading the job postings and counting them…
AI Career Compass · role
Makes AI run fast on hardware: chips, GPU kernels and compilers. Chip design and verification, GPU and accelerator software, kernels, compilers and performance engineering.
Measured In our tracked sample, roles in AI chips, kernels and compilers make up 57 of 1,694 AI postings, or 3%.1
Measured The skills these postings ask for most are Python (asked for in 28 of 57), C++ (21) and chip design and accelerators (19).123
Written by an AI analyst (claude-opus-5-5) and checked by code against the numbers it cites.
What these postings say the work involves, counted from their lists of duties. A posting can list several.
Out of these 57 postings. “Asked for” adds required, one of several options, and listed; preferred is counted apart.
Programming, data, maths and statistics, and domain knowledge
| Skill | Asked for | Required | One of several | Listed | Preferred |
|---|---|---|---|---|---|
| Python | 28 of 57 | 10 | 5 | 13 | 4 |
| C++ | 21 of 57 | 4 | 5 | 12 | 5 |
| Software engineering practice | 10 of 57 | 1 | 0 | 9 | 5 |
| Mathematics | 7 of 57 | 1 | 1 | 5 | 0 |
| Rust | 6 of 57 | 1 | 3 | 2 | 1 |
| Algorithms and data structures | 5 of 57 | 3 | 0 | 2 | 4 |
| C | 5 of 57 | 1 | 2 | 2 | 2 |
| Data analysis | 4 of 57 | 0 | 0 | 4 | 0 |
Modelling, training, retrieval, agents, evaluation and other methods
| Skill | Asked for | Required | One of several | Listed | Preferred |
|---|---|---|---|---|---|
| Machine learning | 18 of 57 | 7 | 0 | 11 | 18 |
| Simulation | 9 of 57 | 3 | 0 | 6 | 3 |
| Large language models | 6 of 57 | 1 | 2 | 3 | 10 |
| Deep learning | 6 of 57 | 1 | 2 | 3 | 10 |
| Evaluation (evals) | 5 of 57 | 1 | 0 | 4 | 5 |
| PyTorch | 4 of 57 | 0 | 4 | 0 | 8 |
| JAX | 4 of 57 | 0 | 4 | 0 | 3 |
| AI agents | 4 of 57 | 1 | 0 | 3 | 1 |
Deployment, infrastructure, reliability, security, monitoring and cost
| Skill | Asked for | Required | One of several | Listed | Preferred |
|---|---|---|---|---|---|
| Chip design and accelerators | 19 of 57 | 10 | 0 | 9 | 8 |
| Performance and latency | 16 of 57 | 5 | 0 | 11 | 5 |
| GPU programming (CUDA, kernels) | 13 of 57 | 4 | 0 | 9 | 10 |
| ML compilers | 10 of 57 | 3 | 0 | 7 | 8 |
| HPC and GPU clusters | 9 of 57 | 4 | 0 | 5 | 13 |
| System design and architecture | 8 of 57 | 2 | 0 | 6 | 0 |
| Distributed systems | 6 of 57 | 1 | 0 | 5 | 4 |
| MLOps and ML platforms | 6 of 57 | 2 | 0 | 4 | 3 |
Publishing and leading research
| Skill | Asked for | Required | One of several | Listed | Preferred |
|---|---|---|---|---|---|
| Open-source contributions | 1 of 57 | 0 | 0 | 1 | 6 |
| Research publications | 1 of 57 | 0 | 0 | 1 | 3 |
Customers, communication, product thinking and leadership
| Skill | Asked for | Required | One of several | Listed | Preferred |
|---|---|---|---|---|---|
| Communication | 9 of 57 | 2 | 0 | 7 | 3 |
| Leading and mentoring | 6 of 57 | 4 | 0 | 2 | 1 |
| Technical writing | 4 of 57 | 2 | 0 | 2 | 0 |
Measured Machine learning knowledge is listed as preferred in 18 of 57 postings we track.1
Measured Experience with high-performance computing (HPC) and GPU clusters, large groups of computers that work together, is listed as preferred in 13 of 57.1
Employers list these as preferred. That they ask does not show that these skills decide who is hired.
The systems these postings expect people to build, run or work within.
Only what the postings state. Most do not say everything, so each number has its own count of postings.
28 of 57 postings state a minimum; the median is 5 years.
30 of 57 postings state a degree; 19 of them accept equivalent experience instead.
4 more postings list a PhD as preferred.
3 postings are remote. Of these, 2 name where you must live or work from, 1 does not say, and 0 say anywhere.
What employers advertise, grouped only when the ranges are comparable. Not a prediction of what you would earn.
Measured In the postings we track, the median of advertised range midpoints is 346,500 USD per year, base pay, for full-time roles in the US, based on 18 comparable pay disclosures.12
45 of 57 postings (79%) advertise a pay range. Employers that publish ranges may differ from those that do not, and some places require it by law, so these ranges do not describe all pay.
| Group of alike ranges | Postings | Median of range midpoints | Middle half of midpoints |
|---|---|---|---|
| USD, a year, base pay, United States, full-time3 employers | 18 | $346,500 | $283,125 to $396,000 |
This is advertised pay, not what anyone earns. We pool ranges only when currency, pay period, pay type, country and employment type are the same, and only with at least 10 of them. Nothing is converted. 27 other ranges could not be pooled (period not stated, several countries, or too few alike).
Advice built from what these 57 postings ask for. Tick what you can already do; we suggest up to three steps and one portfolio project, each with the requirement behind it and a way to show the result.
Analysis We began tracking on 6 October 2026, so the 57 postings here come from one reading so far and cannot show a trend yet.1
Measured The postings we track come from 12 employers that publish their jobs through public feeds, not from the whole job market.1
Measured Many open listings are not new: only 9 of 57 were first published in the 30 days before the data cutoff, and 6 of 57 more than a year before it.12
How everything is counted: our method. All roles: AI Career Compass.