Design and build a production-grade AI memory and retrieval engine for personalized, privacy-preserving LLM experiences. Responsibilities include developing multi-tier storage and hybrid retrieval pipelines, memory extraction and classification, adaptive updates, evaluation frameworks, and reliability safeguards. The role researches advanced memory architectures, supports agentic workflows and tool integrations, collaborates with product management, and owns production operations. Strong software engineering, Python, NLP/ML, vector database, LLM orchestration, privacy, and RAG expertise are required.

Design and own backend infrastructure for a privacy-first AI memory platform. Responsibilities include multi-tier memory storage, secure multi-tenant isolation, real-time APIs, caching, LLM integrations, retrieval systems, synchronization, monitoring, deployment pipelines, and production reliability. The role requires close collaboration with ML engineering and product teams while owning architecture, infrastructure, testing, security, and operational troubleshooting.

Design and build production-grade AI memory systems, including retrieval pipelines, storage tiers, privacy filtering, memory extraction, adaptive updates, hybrid search, and evaluation frameworks. Ship and maintain LLM-powered features, diagnose RAG and prompt failures, implement agentic workflows and tool integrations, and ensure reliable degradation and fallback strategies. Collaborate with product management to define feasible memory-system capabilities while owning production reliability.