I design how AI agents and the teams around them remember — knowledge graphs, retrieval, temporal and verifiable data, and the identity and trust between agents — and I have shipped it in production, not just on slides.
Engineering since 2002. Founding engineer twice over — at Affinidi, on self-sovereign identity, and at Mykin, on a privacy-first personal AI with its memory on the device. Before that, architecture at Zalando scale.
Founding engineer and principal architect. Shaped the technical vision with the CTO and built on-device semantic memory for a privacy-first personal AI — a personal knowledge graph with vector search and clustering, running on a phone. Partnered with Turso to bring vector search to mobile, and represented the architecture in a $2M round.
Founding engineer, architecture group and Office of CTO. Led cross-company architecture reviews, introduced ADRs and C4 across engineering, and took a COVID verifiable-credentials platform from zero to a live airport integration in four weeks.
Architected and led Transaction Core, the order backend for 40M+ customers; led the checkout monolith's move to microservices and co-ordinated 18 teams, about 100 engineers, for the loyalty-programme rollout.
I believe true leaders build teams, not software — the teams build the software. Today those are hybrid teams of people and agents that shape each other, and a team is only as good as what it can remember, share and trust. That is why I work on memory.
Off the clock: a tea-drinking architect who writes children's books.
Do, validate, iterate, learn, teach. Extend yourself with people and tools.
Architecture consulting on agent memory, knowledge graphs, and agent identity and trust: design partnership, architecture reviews and team workshops — the same problems the books are about, on your system.
Agentic memory architectureFractional architectAgent identity & trustArchitecture reviewWorkshops & training
Explore the services18 published books on agent memory, knowledge modelling, the graph databases memory runs on, and how architecture turns into business impact — with a guide to which to read in depth and which to skim.
AI Agents Memory Empowered by Knowledge GraphsBeyond Context GraphsTemporal Aware AI MemoryGraph Engineering: The Loop That Remembers …and 14 more
Browse the catalogueA shared space for agents and humans.
Turns Markdown notes and code into a typed, validated knowledge graph that people and AI agents read and write together. A shared space where people and their agents work from the same graph.
A graph database small enough to live where the agent lives.
An embeddable RDF / SPARQL graph database, built as the knowledge and memory layer for agentic AI running at the edge.
Layered, never-forget memory for agents.
An embedded graph database on SQLite in which every fact has an id, so facts carry provenance and belief — and every change is kept with when it was made and when it was true.
All memory projects → Independent research behind them: Metagraph Encodings over Identified Triples · Layered Bitemporal Graphs
Or write to me: pavlyshyn@gmail.com