Platform Engineer
Layer10
full-remotemidpermanentbackenddata United States Yesterday via LinkedIn
220,000 - 250,000 USD/annual
See how well this job matches your profile
Sign up to get an AI match score and generate a tailored application in seconds.
Get your match scoreTags
Vector DatabasesInformation RetrievalRetrieval PipelinesEmbeddingsPythonTypeScriptObservabilityLLM ToolingAgent SDKsKnowledge Graphs
About the role
Role Overview
Platform Engineer (Memory) at Layer10, focused on building production memory infrastructure that agent deployments rely on daily. You’ll design systems for durable, queryable, and contextually rich memory spanning vector stores, relational models, and retrieval architectures.
Responsibilities
- Design and build memory subsystems for agent recall, context assembly, and long-term learning across deployments
- Implement and optimize vector database usage (storage, indexing, and retrieval at scale)
- Build and iterate on retrieval pipelines (embedding generation through re-ranking) to serve high-quality agent context in real time
- Develop relational memory models for structured agent state, session history, and cross-agent knowledge sharing
- Integrate memory into agent harnesses and SDKs (e.g., Claude Agent SDK, Codex, LangGraph)
- Add observability for memory quality, latency, and cost trade-offs
- Collaborate with the founding team to shape memory architecture and the product roadmap
Requirements
- 3–5 years of software engineering experience, including meaningful time on data-intensive or ML-adjacent systems
- Hands-on production experience with vector databases and/or embedding pipelines
- Strong fundamentals in Python and/or TypeScript; comfortable working across the stack when needed
- Experience building with at least one agent SDK or orchestration layer (agent frameworks/LLM tooling)
- Strong intuition for information retrieval (precision, recall, latency, relevance)
- Ability to thrive in an early-stage environment with broad scope and fast-changing context
Bonus Points
- Experience with graph databases/knowledge graphs (e.g., Neo4j)
- Background in search infrastructure (e.g., Elasticsearch/Solr) or recommendation systems
- Contributions to open-source agent or memory tooling
- Deep opinions on what current “agent memory” approaches get wrong
Compensation & Work Style
- Fully remote and async-friendly
- Base salary: USD 220K–250K, commensurate with experience, plus equity
About Layer10
Layer10 is building an agent deployment platform that uses persistent, structured memory as a core primitive. The platform enables teams to deploy agents that can remember, learn, and evolve over time, with production infrastructure for agent runtime memory systems.
Scraped 4/22/2026