Robotics and RL Researcher (Research Coding Tasks)

$80/hr

rate as listed

Workload: ~15h/week
Engagement: Project-based
Location: Worldwide
Field: Engineering

Convert published robotics and reinforcement learning research into reproducible coding tasks for frontier AI agents. The work includes benchmark design, hidden evaluation, reference implementations and review of agent submissions.

The work

Start with a recent robotics or reinforcement learning paper that has released code, ideally one of your own. Remove the implementation of its core contribution, package the remaining repository as an agent-solvable task, write the task statement and build a hidden reference solution. You will also create a second solution that exceeds the required benchmark bar, then review later agent attempts.

What the platform screens for

The screen focuses on a verifiable publication and code-release record, PhD status and depth in robot learning, sim-to-real transfer, control and reinforcement learning. Expect follow-up questions about exactly what you implemented, how a reported improvement was measured and how you would define held-out workloads and per-metric pass thresholds without leaking the target solution. First-author work, Docker experience and reproducible benchmark design are preferred. Background checks are part of the application process.

Logistics

This is fully remote, asynchronous contract work with flexible scheduling. Each task typically takes a few hours, though the exact effort depends on the paper, repository and evaluation setup.

Pay band: $80/hr

What they look for

Published robotics and reinforcement learning research
Released code for at least one published paper
PhD in computer science or a related field, completed or in progress
Robot learning, sim-to-real transfer, control or reinforcement learning depth
Research code implementation and benchmark evaluation

Included with this listing

Interview prep — 10 questions

Matched to what this platform's screening actually measures. Answer each out loud in 60–90 seconds: claim → concrete example → implication.

What it tests: This verifies that the candidate can connect a specific publication, public code artifact and personal implementation role rather than relying on a general research summary.
A strong answer: Name the paper, venue, repository or artifact link, authorship position and the components you personally designed or implemented. Distinguish your contribution from work completed by collaborators.
Where people lose it: Citing a paper without released code, giving an inaccessible repository or claiming ownership of the full implementation when responsibilities were shared.

Summary, eligibility guidance and screening preparation are prepared by Sidequest from the platform's posting. Check the source listing before applying.