Abby can now orchestrate complex multi-step research workflows with independent steps running in parallel. High-risk tools (modify concept sets, update cohort criteria, execute SQL) join the toolkit with safety validation. Dry run mode simulates actions before execution. Workflow templates encode OHDSI best practices into one-click study designs.
Abby can now take actions, not just answer questions. "Build me a diabetes cohort" generates a reviewable multi-step plan — create concept sets, define the cohort, generate the patient count — that executes with one click after user approval. Every action is logged with checkpoint data for rollback. Phase 4 adds supervised autonomy with safety rails.
In a single development session, we shipped three phases of a cognitive architecture that transforms Abby from a stateless RAG chatbot into a persistent, intelligent, context-aware research assistant. She now remembers who you are, routes complex questions to a more powerful brain, traverses clinical concept hierarchies, and warns you when your data has gaps. This post tells the complete story — the problems we solved, the architecture we built, and the engineering decisions behind 188 passing tests across 60+ new files.
Abby now learns from the entire research community. When a researcher builds a successful diabetes cohort, that pattern becomes available to every other researcher. Questions asked three or more times across users automatically become institutional FAQs with vetted answers. Data quality findings discovered by one team warn all teams. Phase 6 completes the Abby 2.0 cognitive architecture.
Abby now has two brains. Simple queries stay local on MedGemma (fast, free). Complex reasoning escalates to Claude via API (powerful, cloud). A PHI sanitizer blocks any patient data from leaving the network, and a cost tracker with circuit breaker keeps spending within budget. Researchers get smarter answers on hard questions without compromising privacy or breaking the bank.
Abby now understands that metformin is a drug used for Type 2 diabetes mellitus, which is a subtype of diabetes mellitus. She traverses OMOP concept hierarchies, finds siblings and related concepts, and warns researchers when they're building on sparse data. Phase 3 turns keyword matching into relational understanding.
Abby now builds a persistent research profile for every user, tracks conversation topics across turns, and assembles context through a ranked, budget-aware pipeline. Phase 1 of the Abby 2.0 cognitive architecture lays the memory foundation — moving from stateless Q&A to a personalized research assistant that gets better with every interaction.
What started as "Abby's responses are slow" turned into an 18-hour deep dive that touched every layer of the AI stack — from GPU driver backends to embedding model race conditions to the fundamental question of why a 4-billion-parameter medical LLM was confidently inventing researcher names. By the end, Abby went from 15-25 second hallucinated responses to 2-5 second grounded answers backed by 167,000 vectors of medical knowledge — and we found that an 8-megabyte systemd memory lock was silently killing 25% of all GPU inference requests.
Abby can now answer "What concept sets do we have for diabetes?" and "How many patients are in our CDM?" with real data — queried live from the Parthenon PostgreSQL database at response time. Eight contextual tools give her awareness of concept sets, cohort definitions, vocabulary concepts, Achilles characterization stats, data quality results, cohort generation counts, CDM summaries, and analysis executions.
Today we transformed Abby from a capable AI assistant into an OHDSI domain expert backed by the largest curated outcomes research knowledge base we're aware of in any open-source platform. By the end of the day, Abby's ohdsi_papers ChromaDB collection held 79,070 SapBERT-embedded vectors spanning peer-reviewed research papers, the Book of OHDSI, HADES package documentation, and a decade of practitioner Q&A from the OHDSI forums.