Job Title

About Particle Execution:We’re an AI solutions specialist and we’re building modular AI infrastructure for execution - including a second brain for our clients.Second brain is a scalable knowledge and content-generation system that supports high-context AI agents for all our clients. It stores, learns from, and reasons with everything your business knows - enabling execution through strategy, content, and conversation.We’re now looking for an engineer who can work with the rest of our team and can help us turn per-client data into a persistent, intelligent memory system.The Role:We’re hiring a Memory Systems Engineer to design and build the backbone of our product’s long-term memory - initially for content automation, eventually for agentic workflows across industries.This is a hands-on role for someone who can ship fast, but also think deeply about system design. You’ll work closely with the team to architect, test, and evolve how memory works across vector search, feedback integration, and contextual retrieval.What You’ll Build:A modular, per-client memory system that supports:Dense vector retrieval (FAISS, Qdrant, Weaviate)Feedback-aware updates based on edits, ratings, and post performanceDynamic context routing based on generation type (e.g. strategy vs post)A scalable infrastructure to support:Dozens → thousands of clientsLow-latency retrieval and context injection for LLM generation(Optional but welcome) Foundations for:Graph-based or symbolic memory layersTree-of-thought traceabilityAgent memory logs and replayable reasoning chainsYou Might Be a Fit If You:Have built or contributed to LLM-integrated apps or systemsKnow how to build RAG pipelines from scratch (chunking, embedding, querying)Have worked with vector databases (FAISS, Qdrant, Pinecone, etc.)Have strong opinions on how memory should evolve over timeCare about building real-time, adaptive, and traceable AI systemsThink pragmatically now, but architect for future generalizationBonus Points For:Experience with LangChain, LlamaIndex, or LangGraphFamiliarity with graph databases (Neo4j, Dgraph) or key-value memory systems (Redis, DuckDB)Knowledge of hyperdimensional memory or symbolic-connectionist hybridsA portfolio or writeup of RAG or memory system designWhat You’ll Get:Ownership over core product infrastructureA chance to influence how next-gen AI agents think and rememberA team that moves fast, thinks big, and builds practicallyCompetitive compA growing customer base and a roadmap full of execution challengesWe’re not just building AI that chats. We’re building AI that remembers, improves, and executes. If you think memory is the key to agentic intelligence - we want to talk.