Backend Engineer

Mem0 is building the foundational memory layer for AI agents, enabling them to learn, remember conversations, and build context over time. As one of the first two engineering hires, you will own the backend platform, setting the technical standard and collaborating closely with research and frontend teams to deliver fast, reliable features that power millions of AI interactions.

What you'll do

• Design and ship robust REST APIs, including defining contracts, versioning, authentication, rate limits, migrations, and documentation.
• Model data and schemas across relational (PostgreSQL) and graph (Neo4j or equivalent) stores, ensuring data integrity and performance.
• Debug customer issues end-to-end by tracing with logs, metrics, and traces, reproducing problems, shipping fixes, and implementing preventative guardrails.
• Optimize system performance by tuning slow SQL queries, utilizing indexes, partitioning, pagination, and caching with Redis.
• Build services in Python, leveraging async programming with frameworks like FastAPI, Starlette, Django, DRF, or Flask, alongside background jobs, queues, and schedulers.
• Operate services in the cloud by containerizing with Docker, deploying on Kubernetes (EKS), and utilizing AWS primitives such as EC2, RDS, Aurora, S3, and IAM.
• Instrument all systems with custom metrics, structured logging, and tracing, setting clear SLOs and alerts using tools like CloudWatch, Prometheus, and OpenTelemetry.
• Collaborate effectively with frontend and research teams to scope APIs and ensure seamless delivery of features to production.

What Mem0 is looking for

• 5 to 8 years of experience building and shipping backend systems with REST APIs to production.
• Strong Python fundamentals, including experience with async programming and a major web framework (FastAPI, Django, or Flask).
• Solid data modeling and SQL skills, with hands-on experience in query tuning and performance debugging in PostgreSQL or MySQL.
• Experience with graph databases such as Neo4j or Amazon Neptune, understanding appropriate data modeling trade-offs.
• Comfort operating services on AWS using Docker and Kubernetes.
• Demonstrated ability to perform root-cause analysis and take ownership from incident resolution to prevention.
• Excellent communication and collaboration skills, working effectively with frontend, research, and customers.
• Willingness to work fully in-person, 5 to 6 days a week, from Mem0's San Francisco office.
• Bonus points for experience with GraphQL or gRPC, event-driven systems (SNS, SQS, Kafka), background workers (Celery, RQ), caching, rate limiting, multi-tenancy, feature-flag strategies, security/privacy best practices, deep observability (OpenTelemetry, SLO-based alerting), prior work with search/retrieval/memory systems, and on-call experience.