Hi, I'm Ramona, an Independent Researcher, Solutions Architect and Educator. I bring over 20 years of rigorous database engineering experience to the world of AI; my current focus is on System Integrity—applying the strict determinism of data engineering to the probabilistic nature of AI Agents. More to the point, I build instruments that measure whether database and AI agents do, and then publish the experiment alongside the result. In that way, you can reproduce it or show me where I'm wrong. This site is built the same way: generated from data, and the build fails on an unknown tag, a broken anchor, or an unregistered number.
Portfolio · Writing · LinkedIn · ORCID
sql-benchmarks-dagster · pip install sqlbenchdag · MCP server in the official registry
Every experiment is a capsule addressed by an 8-character SHA-256 fingerprint over its config, SQL, and every line of measurement-relevant Python. Change the method, the ID changes. Cold-cache execution, hard row-count assertions, an integrity seal on every capsule — and OpenTimestamps proofs anchored to Bitcoin on the four published Quack capsules.
| Finding | Capsule |
|---|---|
| DuckDB-over-Quack (pushdown) beats PostgreSQL at every scale past the noise floor — 4.4× @100K, 6.1× @1M, 13.2× @10M rows | 902d1277 |
| Attach-mode overhead grows with scan size (2.6× @100K → 9.5× @10M); pushdown stays flat at ~2× | b8e2bfaf |
| That flat ~2× residual is reduced server-side parallelism, not protocol transport | 25b0e134 |
| Attach mode cannot execute multi-table joins at all; pushdown holds ~1.8× on a 3-way TPC-H join | b198363e |
Full published-capsule index →
| rctruta.github.io | The repository powering this website is built as a data modeling project. → |
| lakehouse-semantics | A suite of 23 correctness probes executed across DuckDB, DuckDB+Delta, PostgreSQL, and Databricks SQL. → |
| harness-bench | I engineered harness-bench to evaluate agent deployments before they reach production. → |
| malloy-publisher-agent-study | The Malloy Publisher ships agent skills alongside its MCP server. → |
| bauplan-agent-parse-study | Per-turn agent traces against a commercial lakehouse platform. Surfaced three defects I reported upstream — #369, #370, #371 — including an error message that actively sends an agent in the wrong direction. → |
| agent-telemetry | A local analytics package that parses raw JSONL session transcripts from coding agents (Claude Code) into un-opinionated process metrics: tool_calls_per_turn, reads_before_first_edit, repeated_commands, error_rate, seconds_to_first_write, and edit_revisits. → |
| podcast-rag | Ask a question about a podcast, get an answer and a link to the exact second someone said it. LanceDB, local-first, no cluster and no API key. Built a labelled eval set and found it answering 17 of 20 questions its sources couldn't support; recalibrated precision 0.56 → 0.87. → |
| music_recommendation_system | Top-10 song recommendation over the Million Song Dataset Taste Profile Subset: two million listening events, ten thousand songs. → |
| adversarial-judgement-research | Metatracing engine and execution traces for structural failure modes in multi-agent LLM consensus pipelines: status bias, persona bleed, frame break, axiomatic refusal, consensus contagion, semantic camouflage. → |
| ai-security-testbed | A deterministic matrix engine for adversarial audit of multi-agent pipelines: temperature 0, replication, status and thinking-budget sweeps, rule-based grading. 416 traces across frontier and local models. → |
| ai-agent-utils | Boilerplate and security guidelines for collaborating safely with autonomous coding agents, with the gates already enforced rather than written down and hoped for. → |
- A README is a Set of Falsifiable Claims — Testing mine found a security command that never worked.
- Measuring Quack: DuckDB's New Client-Server Protocol — Pushdown holds a bounded ~2x overhead and beats PostgreSQL on analytical aggregations; attach mode does not scale, and cannot join at all.
- The Job That Wasn't — When you apply for a gig job in AI, you might be the product — not the candidate.
- The Emotional P&L — Why we pay compounding interest on unexamined feelings.
- Context is Empathy: The Cognitive Firewalls We Build Under Pressure — Why ignoring structural vulnerabilities in AI isn't ignorance — it's a survival mechanism.
- Neo4j NODES — Structure Is Not Security: Poisoning Graph-Based Agent Memory Through the Extraction Pipeline · virtual · November 12, 2026
- Data in the D — Measure What Matters · Detroit · October 16, 2026
- Canadian Women in Cybersecurity — Consensus Contagion: Status Bias and the RLHF Tax in Agentic AI Routing Nodes · Toronto · May 26, 2026
- AI Tinkerers Toronto — Semantic Laundering: Agentic Memory · Toronto · February 26, 2026