Member of Technical Staff
dmodel
Member of Technical Staff
dmodel
AI interpretability research lab and partnerships
d_model is a fundamental AI research lab partnering with frontier labs to turn their models into capable interpretability and alignment researchers. Alongside our partnerships, we aim to use the agents we build for independent research.
Our team brings experience from places including OpenAI, Google, Anthropic, EleutherAI, and MATS.
As a Member of Technical Staff, you'll build the reinforcement learning environments and evaluations we use to study how AI agents approach alignment and interpretability problems. This role is a strong fit for someone who wants hands-on experience in AI safety, interpretability, and reinforcement learning.
Build reinforcement learning environments and evals to test how AI agents approach alignment problems
Develop graders robust to specification gaming
Find novel interpretability and alignment techniques
Present findings in team research meetings and contribute to papers in ML
Run experiments and identify patterns in model behavior and failure modes
Investigate AI Interpretability techniques within our reinforcement learning environments
Have strong proficiency in Python and ML frameworks (PyTorch or JAX)
Are able to iterate quickly and collaboratively
Are familiar with the alignment and/or interpretability literature
Can generate research ideas in the field and implement others' ideas
Published or presented research in AI safety, especially Mechanistic Interpretability
Graduate degree in physics or mathematics, or equivalent research experience
Academic research experience
Experience post-training language models
Health, vision, and dental benefits
Flexible PTO policy
Catered breakfast and lunch
Monthly workspace and wellness allowances
Commuter benefits
Free SFMOMA admission and group subscriptions to Works in Progress, The New Yorker, Eye, and Aesthetica
Location Policy: Currently, we expect all staff to be in office 5 days a week in our San Francisco location.
Visa Sponsorship: We sponsor visas. Not every role and candidate combination works out, but when we make an offer we commit to a genuine effort, supported by an immigration counsel we work with directly.
Compensation: $150,000-$225,000 + bonus + equity
We encourage you to apply even if you don't believe you meet every single qualification.
d model (dmodel.ai) is an AI research lab that partners with frontier labs to advance interpretability and alignment research. It develops agents and tools to study how to make AI models safer and easier to steer, focusing on independent interpretability research rather than exploiting rewards. The company shares RL environments for open-ended interpretability tasks and publishes research on concept-editing algorithms discovered by large language model agents. It also seeks researchers to work on interpretability projects that teach agents real research skills rather than reward hacks. The business model centers on partnership-driven AI interpretability R&D with frontier lab collaborators. Overall goal: improve AI safety and controllability by building practical interpretability methods, environments, and agents through collaborative research.
Explore more opportunities at dmodel