Member of Technical Staff

dmodel
San Francisco
Workplace: OnsiteFull timeUSD 150,000 - 225,000Function: Software EngineeringSkills: ["Collaboration","Iteration","Research ideation","Pattern identification"]

Build reinforcement learning environments and evaluation pipelines to study how AI agents approach alignment problems. You’ll develop robust graders to prevent specification gaming, run experiments, and analyze model behavior and failure modes. Work closely with researchers to find novel interpretability and alignment techniques, present results in team research meetings, and contribute to ML papers using Python with PyTorch or JAX.

Loading

Loading job details...

Preparing the role view and application actions.

FursaFursa
dmodel
dmodel
1 day ago

Member of Technical Staff

✓ Verified Job

Canonical indexed version, validated from employer's careers page.

Source: Company careers pageValidated by: Fursa AI
Last checked: 14 minutes agoStatus: Live

Job Summary

Build reinforcement learning environments and evaluation pipelines to study how AI agents approach alignment problems. You’ll develop robust graders to prevent specification gaming, run experiments, and analyze model behavior and failure modes. Work closely with researchers to find novel interpretability and alignment techniques, present results in team research meetings, and contribute to ML papers using Python with PyTorch or JAX.
Location: San Francisco
Workplace: Onsite
Employment Type: Full time
Job Function: Software Engineering

Key Responsibilities

  • •Build reinforcement learning environments and evaluations to test how AI agents approach alignment problems.
  • •Develop robust graders to reduce specification gaming.
  • •Investigate and apply interpretability techniques within reinforcement learning environments.
  • •Run experiments and identify patterns in model behavior and failure modes.
  • •Present findings in team research meetings and contribute to papers in ML.

Pay and Benefits

Salary: USD 150,000 - 225,000
Equity and Bonus:Equity
Perks:Health InsuranceDentalVisionRemote Work

Key Requirements

  • •Strong proficiency in Python and ML frameworks (PyTorch or JAX).
  • •Familiarity with alignment and/or interpretability literature.
  • •Ability to iterate quickly and collaboratively in a research environment.
  • •Can generate research ideas and implement others' ideas.
  • •Graduate degree in physics or mathematics, or equivalent research experience. (Strong candidates may also have published/presented AI safety research.)
Experience:AI safetyReinforcement learningMechanistic interpretabilityInterpretabilityML researchMachine learning
Skills:CollaborationIterationResearch ideationPattern identification
Tech Stack:PythonPyTorchJAX

Company Brief

dmodel
Builds AI-driven simulation and digital modeling tools to create realistic synthetic environments and datasets for testing, validating, and improving machine learning models and autonomous systems across industries.
Industry: AI & Machine Learning
Website