The Research Bench
Not everything is a polished product.
My laboratory for agent protocols, memory graphs, and experimental tools. The messy experiments here are where my production architectures get proven.
Active Experiments
3 in testingFieldoEXPERIMENT
Experiments · experiment · field opsField-data capture experiment for physical work.
Next.jsPostgres
Inspect experiment →RavenEXPERIMENT
Agent Systems · experiment · ops agentsAutonomous ops agent bench: memory, tools, and fallback.
HermesOpenClawSQLite
Inspect experiment →Graph MemoryEXPERIMENT
Agent Systems · experiment · memory graphsRelationship memory between agents, tools, and knowledge.
PostgrespgvectorHermes
Inspect experiment →Decision Bench
How I chooseArchitectural Tradeoffs
These are the weighted heuristic rules I use when choosing frameworks, model routing, and storage layers:
DATABASE
Postgres
Supabase
Firebase
AGENT FRAMEWORK
OpenClaw
LangChain
AutoGPT