Prateek Chhikara.

Applied AI Engineer at Mistral AI · San Francisco
I turn real-world problems into AI systems that work beyond the demo.
Previously, I was a founding AI engineer at Mem0 and a researcher at USC’s Information Sciences Institute. I’m interested in how we make models more reliable, more perceptive, and better at remembering what matters.
Selected research
All publicationsMind the confidence gap: Overconfidence, calibration, and distractor effects in Large Language Models
Understanding and reducing overconfidence in large language models.
Mem0: Building Production-Ready AI Agents with Scalable Long-Term Memory
Scalable long-term memory for production AI agents.
MLLMs Know Where to Look: Training-free Perception of Small Visual Details with Multimodal LLMs
Helping multimodal models perceive small visual details without retraining.
Latest writing
All writingOutside the terminal
I draw anime-inspired artwork and write about life, research, and the things I learn along the way. Visit the sketchbook