engram is an engineering intelligence platform. it ingests your github repos and notion workspaces, then answers grounded in the org knowledge that actually shipped. cross-validation is coming next; today, cited retrieval is the core.
eu-west-1 times out because the celery worker pulls the embedding model on first boot [1]. the deploy script doesn't warm the cache before health-check[2]. the runbook in notion calls for a 90s grace period; current k8s manifest sets 30s [3]— that's the mismatch.# k8s/eu-west.yaml — patch grace period readinessProbe: httpGet: { path: /healthz, port: 8000 } initialDelaySeconds: 90 # was 30 periodSeconds: 10
ask in natural language, get answers cited to specific files and notion pages. intent detection routes between explain and generate paths.
github oauth + notion oauth. periodic re-index via celery. one-click connect, progress in-line.
browse indexed sources, inspect metadata, and revisit what has been added to workspace memory.
watch celery queues, inspect failed re-indexes, retry with one click. operator-grade observability.
save any answer as a durable artifact — code snippet, summary, or report. durable and citeable inside the product.
compare related code and docs side-by-side today. automated cross-validation and staleness scoring are on the roadmap.
auth, integrations, model preferences, mcp tokens. per-conversation provider overrides on the roadmap.
same retrieval, same answers — exposed via the model context protocol. plug engram into claude code, cursor, vs code, and claude desktop.
┌─────────────────┐ ┌─────────────────┐ ┌─────────────────┐ │ github repos │ │ notion pages │ │ saved outputs │ │ oauth · index │ │ oauth · index │ │ cited answers │ └────────┬────────┘ └────────┬────────┘ └────────┬────────┘ │ │ │ └─────────────┬─────────────┘ │ │ │ [ celery ingest ] ──── chunk · embed · upsert ───── │ │ │ ▼ │ ┌─────────────────────────────────────┐ │ │ postgres + pgvector │ ◄── code is the truth │ chunks · embeddings · meta │ └─────────────────┬───────────────────┘ │ [ retriever ] ─── topK + intent shaping │ ▼ pluggable llm grounded generation │ ▼ ┌─────────────────┴───────────────────┐ │ web ui │ mcp plugin │ api │ └─────────────────────────────────────┘
engram is a thin, opinionated stack assembled from boring components and one practical idea — an engineering memory layer should cite the code and docs it used instead of guessing from model memory.
the chat engine detects intent (explain vs. generate) and adjusts the prompt accordingly; the retrieval pipeline is shared. all llm calls route through a single provider abstraction so we can swap models without touching routes.
start with one github repo and one notion workspace. engram indexes both and gives you cited answers from the sources your team selected.