Skip to main content
Systems Online

I design products and build the AI systems behind them

A decade shipping products for Wegmans, Disney, and AB InBev—now building AI infrastructure that runs in production 24/7. Cortex, my self-healing memory platform, serves 1,600+ memories with dual-path search across a three-node network—all local, zero cloud.

Flagship Project

Cortex

Cortex is a self-healing memory platform for AI agents. Memories go in raw and self-organize — auto-embedded, auto-linked through cosine similarity, auto-clustered by project. Dual-path search merges Postgres full-text and ChromaDB semantic results, ranked with recency bonus. A-MAC admission control quality-gates every write. Adaptive retrieval (Agentic RAG via LangGraph) reformulates queries when relevance is low. Runs natively on Apple M4 Silicon with embedded ChromaDB (PersistentClient) — no Docker, no zombie containers. Three-tier self-healing: process (LaunchAgent auto-restart), application (graceful degradation, Redis write buffer), infrastructure (warm standby, automated backups). Currently running in production with 1,400+ memories and 14,408+ auto-generated edges across 30 projects. Sub-3ms FTS query latency, 2ms unified search (FTS + semantic + graph) on forge-native deployment across a three-node Tailscale network.

<3ms
Query Latency
1,600+
Memories
14,200+
Auto-Generated Edges
30
Projects
Read Case Study

Interested in working together?

Building a product, scaling a team, or figuring out where AI fits? Start a conversation.