Post-Training Researcher
T4 · Engineering & machine learning$220top rate per hour
Posted by SimRealsimreal.co · feeds SimReal-MLBench
Application To do
Checking…0%
- Sign inOnce for all rolesTo do
- Connect a profileOnce for all rolesTo do
- Verify identityOnce for all rolesTo do
- Join the waitlist for this roleTo do
Signing in, connecting a profile and verifying your identity happen once and count for every role.
About the role
You will design reward functions and then try to break them. This is the work the labs say they are short of: writing a grader, attacking it with a model that is actively trying to cheat, and confirming afterwards that it measured what you intended.
What you will do
- Designing invariant-based reward functions for multi-step tasks
- Running degenerate policies against graders and documenting the bypasses
- Reviewing reward specifications for exploitable slack
- Formalising what a robust verifier looks like in a new domain
Who we are looking for
- Post-training experience: RLHF, RLVR, or reward-model work at a lab or serious research group
- Adversarial mindset: You reach for the shortcut before you reach for the solution
- Systems judgment: Understanding of how a weak reward propagates through a training run
- Rigour: Comfort with variance, pass^k, and contamination checks
Credential
Frontier lab or equivalent.
Pay
Up to $220 an hour for this role. Each role shows its top hourly rate, from up to $40 for generalist review to up to $300 for portfolio managers. Your starting rate depends on your credentials; sustained quality moves you towards the top rate. Joining the waitlist is not an offer of work; rates are agreed in writing before any project starts.
More roles in Engineering & ML
$220 / hour · Post-Training ResearcherApply