Research Engineer

About the Company

Mem0 is building the memory layer for AI agents, providing long-term memory that allows AI to remember conversations and build context over time. Backed by a $24M Series A & Seed, the company powers millions of AI interactions for enterprise customers and a large open-source community.

Responsibilities

• Train models for memory extraction, updates, consolidation, forgetting, and conflict resolution
• Turn research paper ideas into working prototypes and production-ready features
• Build automated evaluation layers including relevance, accuracy, and consistency metrics
• Uncover customer pain points and turn them into testable research hypotheses
• Design APIs and data contracts to ensure SOTA latency, reliability, and cost

Requirements

• Experience in RAG or information retrieval (retrieval, ranking, query understanding)
• Experience with model training and fine-tuning LLMs/encoders
• Strong proficiency in Python
• Deep experience with PyTorch
• Familiarity with vLLM and modern serving frameworks
• Experience building evaluation for complex language or retrieval/generation tasks
• Ability to orchestrate data pipelines for low-latency production environments
• Clear and concise communication skills

Preferred Qualifications

• Publications at venues like NeurIPS, ICML, or ACL
• Experience with privacy-preserving ML (redaction, differential privacy)
• Deep familiarity with memory/retrieval literature
• Expertise with embeddings, vector-DB internals, deduplication, and contradiction detection

Benefits

• Fully covered health, dental, and vision insurance
• Daily catered lunch and dinner in-office
• Flexible PTO
• Equity in an early-stage company
• Regular team happy hours and events
• Top-tier equipment