Building systems that reason and remember
REASON
DSPy optimization, RAG pipelines, agent orchestration. Making LLMs do useful work reliably.
- >Prompt optimization & evaluation
- >Content generation pipelines
- >Autonomous agent workflows
REMEMBER
Postgres-native architectures. pgvector, JSONB, semantic search without separate vector or graph DBs.
- >Vector embeddings in Postgres
- >Event-sourced state & temporal queries
- >JSONB for flexible metadata
What I'm Working On
8-Bit Oracle
ACTIVEAn I-Ching oracle that remembers your journey. Building the memory layer now—events, claims, confidence evolution. The kind of AI that develops a relationship with you over time.
Currently implementing: bi-temporal validity, budget-constrained retrieval, and the “oracle knows you” experience.
Pix
Hackathon WinnerAn autonomous oracle agent on Twitter. Performs I-Ching readings and stores insights on OriginTrail's decentralized knowledge graph. Won Consensus HK 2025.
Content Engine
ACTIVEFull-stack content generation system. DSPy-powered pipelines, MCP agent orchestration, research paper ingestion. All backed by Postgres with pgvector for semantic search.
Processing orchestrator patterns, audit wrappers, content variety scoring, model registry for A/B testing.
Also: ZK proofs for verifiable I-Ching, passkey auth, and various attempts at making machines more human.
Augustin Chan
Before AI, I spent 10 years at Informatica building master data management systems for enterprises across APAC and Europe. Large-scale data architecture, complex integrations, the kind of work where you learn that systems need to be robust before they can be clever.
Now I apply that discipline to AI: prompt optimization with DSPy, production agent orchestration, memory systems that let LLMs maintain context across sessions. The engineering that makes AI useful for users.
BS Cognitive Science (Computation) UC San Diego · Bronx HS of Science
Elsewhere
Talks, interviews, and places I've shown up