LLM Context and Memory Engineer
Build the context engine for a proactive, user-controlled health copilot. Reconcile records, conversations, and uncertainty.
RemoteUSD 252,500/year
Anistratenco is building a longitudinal health platform. Health Assistant brings conversations, records, and goals into a context people can inspect and correct. The underlying systems must support more products as that context grows. This full-time role is remote within the United States.
The mandate
Build the path from documents, attachments, text, and conversations to a temporal health-avatar profile that the user can inspect and correct.
Own extraction, reconciliation, retrieval, proactive context assembly, and evaluation. Preserve provenance, observation time, uncertainty, contradictions, consent, and deletion across the full agent loop.
Show us the work
Send two files: a complete project case study and a proposal for this role. No CV or cover letter. Use public evidence and synthetic examples where needed. AI tools are welcome; the technical decisions, evidence, and proposed execution must withstand review.
Project case study
Document one substantial project from the original problem through launch and subsequent results. Explain the team, your own scope, timeline, constraints, architecture or operating stack, key decisions, implementation, and what changed along the way. Include work you can substantiate, a dated baseline, the measurement method and denominator, and the measured result. Discuss a failed approach or tradeoff. Label estimates and evidence you cannot verify.
Show a complete context, retrieval, or memory system you helped take from architecture to production operation. Detail ingestion, schema, model and open-source components, storage, orchestration, retrieval, evaluation, deployment, and incident handling. Explain your contribution, design revisions, failure cases, and results for quality, latency, cost, and stale or contradictory records.
Role proposal
Infer the current LLM and context stack from public behavior and docs; cite evidence and mark unknowns. Design the end-to-end health-avatar pipeline for documents, attachments, text, and conversations. Specify open-source candidates and tradeoffs for parsing, extraction, temporal storage, reconciliation, retrieval, and agent orchestration. Define exact rules for identity matching, provenance, uncertainty, contradictions, correction, consent, and deletion. Include a system diagram, an avatar-control wireframe, deployment topology, operational runbook, observability, and evaluation cases with quality, latency, cost, and isolation targets. Show how proactive actions use reconciled context without treating inferences as medical facts.