First seen by our source: 3 hours ago
Senior Systems & Hardware Experts - AI Benchmarking
@ Outlier
Canada
SIGNALS
Overview
Outlier is hiring senior systems and hardware engineering professionals to write challenging tasks for an AI benchmark. The work involves turning hard problems from one's own field into Linux terminal challenges that advanced AI agents must solve from scratch. Pay is up to $1,350 per task, with onboarding of about 90 minutes plus a live webinar.
Responsibilities:
- Identify difficult problems in systems or hardware engineering
- Translate problems into Linux terminal tasks
- Design challenges that expose limits of advanced AI agents
- Test each task against three models with three attempts each
Requirements summary:
- At least 9 years of systems or hardware engineering experience
- Completed master's degree or higher
- At least one academic or professional publication
- Comfort working in a Linux terminal
- Residence in a listed country
Tools & skills
Source and dates
Source type not verified · apify-linkedin
- Employer posting date
- Not provided
- First seen by our source
- 2026-11-14
- Source record updated
- 2026-11-14
These dates describe the source record, not a live verification that the employer is still hiring. Check the original posting.
Job details
| Seniority | Senior |
| Experience | 9+ years |
| Hours per week | 40 (estimated) |
| Travel | air: none · ground: none |
| Work setting | Home |
| Position | Mostly sitting |
| Physical demands | physical low · computer high |
| Languages | English |
| Education | Master’s degree (required) |
Summary and details written by AI from the posting; the employer’s posting is the authority.
Outlier
| Field | Value |
|---|---|
| Open jobs on Jobsy | 5 |
Job description
Bring your expertise to the problems today’s most advanced AI agents still struggle to solve.
We’re looking for experienced systems engineering or hardware engineering professionals to develop challenging tasks for an AI benchmark. You’ll draw on problems from your own field and turn them into clearly defined challenges that an AI agent must solve from scratch in a Linux terminal.
What you’ll do
Earn up to $1,350 per task, with opportunities for higher rates based on task quality and submission volume. Tasks take six to seven hours on average, although initial tasks may take longer as you get familiar with the process.
We’re looking for experienced systems engineering or hardware engineering professionals to develop challenging tasks for an AI benchmark. You’ll draw on problems from your own field and turn them into clearly defined challenges that an AI agent must solve from scratch in a Linux terminal.
What you’ll do
- Identify difficult, meaningful problems in your area of expertise.
- Translate those problems into tasks that can be solved in a Linux terminal.
- Design challenges that expose limitations in advanced AI agents. Each task is tested against three leading models, with three attempts per model, and must prove difficult enough that the models mostly fail.
- At least 9 years of relevant professional experience in systems engineering or hardware engineering.
- A completed master’s degree or higher. Candidates with a completed bachelor’s degree, at least 10 years of relevant professional experience and a publication may also be considered.
- Applicants must have completed their degrees and must not currently be enrolled in a degree program.
- At least one academic or professional publication.
- Practical comfort working in a Linux terminal.
- Current residence in Argentina, Mexico, Puerto Rico, New Zealand, Canada, the United Kingdom, the United States, Australia, Peru, Colombia, Austria, Ireland, Spain, Luxembourg, Italy, France, Germany, Belgium, the Netherlands, Sweden, Finland, Singapore, Denmark, Norway and Switzerland.
- Task-based project expected to run for the next few weeks.
- Onboarding includes two courses and a graded assessment, taking approximately 90 minutes.
- A live onboarding webinar is also required before moving to tasking.
Earn up to $1,350 per task, with opportunities for higher rates based on task quality and submission volume. Tasks take six to seven hours on average, although initial tasks may take longer as you get familiar with the process.
Journal
Only you can see this. Kept on this device, and in your account when you sign in.