Research Fellowship - Mechanistic Interpretability

• Develop mechanistic interpretability methods for understanding internal representations, features, circuits, and computations in language models and agents.
• Investigate how model internals can be used to generate intrinsic rewards, auxiliary objectives, diagnostics, or training signals for reinforcement learning.
• Design and run experiments that test whether interpretability-derived signals improve learning, exploration, generalization, robustness, or sample efficiency.
• Compare internally derived rewards against baselines such as human-generated verifiers, reward models, task-level outcome rewards, and standard intrinsic motivation methods.
• Use techniques such as probing, activation analysis, sparse autoencoders, causal interventions, feature attribution, or representation analysis to study model behavior.
• Analyze failure modes, including reward hacking, spurious features, non-causal correlations, objective misspecification, and overfitting to narrow evaluation distributions.
• Build research code, evaluation harnesses, and experimental infrastructure that make results reproducible and useful to the broader team.
• Communicate research progress clearly through written updates, internal presentations, and final project outputs.

• Currently enrolled in a PhD program in machine learning, computer science, artificial intelligence, computational neuroscience, mathematics, or a related technical field. Exceptional candidates with equivalent research experience may also be considered.
• Track record of research excellence or strong research promise, demonstrated through publications, preprints, open-source work, technical projects, competitions, or publicly available artifacts.
• Working understanding of reinforcement learning.
• Familiarity with mechanistic interpretability, representation analysis, or empirical methods for understanding neural networks.
• Strong programming ability in Python and experience with at least one major ML framework such as PyTorch or JAX.
• Clear written and verbal communication of technical ideas.

• Experience with LLM post-training methods
• Familiarity with intrinsic motivation, unsupervised RL, auxiliary objectives, representation learning for RL, or curiosity-driven learning.
• Experience with scalable ML experimentation, distributed training, experiment tracking, or reproducible research infrastructure.
• Interest in turning mechanistic understanding into practical training methods, rather than only analyzing models after training.

• This role is based in our San Francisco office; for exceptional candidates we are willing to consider a hybrid arrangement

Vmax is an applied research lab developing AI capable of open-ended learning. We are building systems to exceed humans in all capacities by optimizing beyond the local maxima of learning from human expertise.