Research Engineer/Scientist, Post-Training

Responsibilities

• Training models for better agentic tool-use, particularly for context management
• Designing mechanisms for continuous model weight updates post-deployment without catastrophic forgetting
• Designing and running experiments to improve understanding of the interplay between data mixtures, training algorithms, and models
• Building infrastructure for generating and collecting synthetic data at scale
• Building challenging evals for measuring agentic capabilities

Requirements

• Proficiency in Python and deep learning frameworks like PyTorch
• Expertise in post-training techniques such as SFT fine-tuning, reinforcement learning, reward models, and preference learning
• Ability to balance execution speed with empirical rigor
• Proven track record of impactful research through breakthrough publications or open-source contributions
• Experience delivering real-world impact beyond pure academic work

About the Company

Founded by the creators of MemGPT from UC Berkeley’s Sky Computing Lab, Letta is building self-improving artificial intelligence. We create agents that continually learn from experience and adapt over time, powering production systems for companies like 11x and Bilt Rewards.