Research Engineer / Scientist, Post-Training

SOLVING SELF-IMPROVING SUPERINTELLIGENCE The human brain is a sponge. Today’s AI brains are brittle and rigid. At Letta, we’re building self-improving artificial intelligence: creating agents that continually learn from experience and adapt over time. Founded by the creators of MemGPT from UC Berkeley’s Sky Computing Lab (the birthplace of Spark and Ray). Backed by Jeff Dean, Clem Delangue, and pioneers across AI infrastructure. Our agents already power production systems at companies like 11x https://www.letta.com/case-studies/11x and Bilt Rewards https://www.letta.com/case-studies/bilt, learning and improving every day. We’re assembling a world-class team of researchers and engineers to solve AI’s hardest problem: making machines that can reason, remember, and learn the way humans do. Note that this role is in-person (no hybrid), 5 days a week in downtown San Francisco. YOUR ROLE You will pioneer post-training techniques that improve how well LLMs can be integrated into complete agentic systems. At Letta, you'll work with a world-class, tight-knit team of AI researchers and engineers towards our vision of self-improving superintelligence. Advance the field through open publishing of research through papers, technical reports, blog posts, and open-source code. WHAT YOU’RE RESPONSIBLE FOR:

• 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 run 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 WHAT WE'RE LOOKING FOR:
• Proficiency in python and deep learning frameworks eg. PyTorch
• Expertise in post-training techniques e.g. SFT fine-tuning, reinforcement learning, reward models, preference learning
• Ability to balance execution speed with empirical rigor
• Proven track record of impactful research (breakthrough publications and/or open-source contributions)
• Real-world impact beyond pure academic work OUR CULTURE SIGNS IT COULD BE A GREAT FIT:
• You want to maximize your impact: you want work on a small, incredibly talented team where every individual plays a huge role in the team's success. You wonder what it would have been like to be at OpenAI when it was just a dozen people, or Google when it was just a couple of grad students in a garage.
• You’re excited to go head-to-head with tech giants, frontier labs, and other startups that are many times our size in both headcount and funding.
• You are fundamentally opposed to closed frontier AI that is controlled by a handful of billionaires and private tech companies. SIGNS IT’S A BAD FIT:
• You like stability, and get stressed out when there’s nobody telling you exactly what to do. We look for people that thrive in ambiguity and can intuit the most important problems by talking to customers and dogfooding our product.
• You want to work a 9-5, and value clear separation of work from life. The stakes are the highest they've ever been, and the only moat is execution velocity. Operating on a strict 9-5 guarantees failure.
• You value titles or want to people-manage. Letta is a flat company where every researcher and engineer is an individual contributor. OUR INTERVIEW PROCESS:
• Initial screen (30 min)
• Technical screen (1-1.5 hours)
• Paid in-person work trial (2 days onsite in SF)

laptop-ats-crawler v2