How a Second Brain Learns to Improve the Company and Itself
Two independent gap detectors, a nightly digest, cost as a behavioural signal, and an honest audit of which feedback loops actually close and which terminate in a table nobody reads.
The Production Engineering Behind an Agent That Never Just Stops
Detached runs, Redis session locks with database fallback, SSE replay buffers, repairing unpaired tool calls after a restart, protocol-safe compaction, and timezone-correct scheduled autonomy.
How to Give AI Agency Without Giving It Reckless Power
Six rungs of agent agency, capability checks re-run at dispatch, mandatory coding plan approval with repository grounding, and why there is deliberately no merge primitive.
Company Memory Is a Governance Problem, Not a Vector Database
Why we shipped governed AI memory with zero embeddings: Postgres tsvector retrieval, OR-lexeme queries, role-scoped visibility enforced in SQL, revision snapshots, and the gaps we have not closed.
An AI Is Only as Intelligent as the Reality It Can Access
How Alfred reaches company data safely: a self-hosted MCP gateway, per-provider path allowlists, lossless response paging, and a Chrome extension with ECDSA P-256 device pairing.
Everyone Wants a Company Second Brain. Almost Nobody Is Building One.
The architecture behind Alfred, DialNexa’s internal AI agent: 45 tools, prompt cache tiers, four-layer read-only enforcement, and why trust has to live in the code rather than the prompt.