About Me
Hi I’m Katrina!
I earned my B.Sc. in Computer Science from the University of Manitoba, where I developed an interest in data science and machine learning. During my degree, I co-authored three peer-reviewed papers published by IEEE that explore how data mining and fuzzy logic can help predict health outcomes.
Most recently, I’ve been working at Dotzlaw Consulting as a software developer and technical writer where we built a general-purpose document assistant that ingests 51,000+ chunks across Markdown, HTML, PDF, and DOCX spanning four repositories. Hybrid vector and BM25 keyword search with cross-encoder re-ranking returns sourced answers in under 3 seconds, and a single iframe-based viewer renders every format without any frontend changes, so the assistant embeds into any host application as a pure backend integration.
I also built a natural language BI platform that turns plain English questions into complete six-card dashboards against a 90.5-million-row production database in under 30 seconds. We shipped two versions of the same AI pipeline: a Metabase-embedded build for clients who want a full BI platform with drill-through, saved dashboards, and scheduled reports, and a native React build using ECharts, AG Grid, and Leaflet for clients embedding AI dashboards inside existing products without a Java runtime. Both reach 100% SQL success across 10+ query categories through a three-layer reliability architecture of T-SQL prompt rules, deterministic post-processing fixes, and a retry safety net.
Alongside the project work, I’ve written 14+ articles covering the infrastructure and tooling behind modern AI systems: vector databases, RAG, graph databases, GraphRAG, and a deep-dive series on building production-grade autonomous coding agents with Claude Code.
I also worked as a freelance developer creating responsive websites, ACID-compliant relational databases, and custom WordPress plugins. Additionally, I’ve taught coding to kids ages 6-18, which enhanced my ability to communicate problems and solutions to individuals of varying skill levels and learning styles.
When I’m not writing code, I like to spend my time watching the Winnipeg Jets and crocheting, hanging out with my dog, and playing video games.
What I Bring to Teams
- Technical Expertise: Expert-level Python development with strong skills in JavaScript, C#, C++, and SQL. Hands-on experience with LLM application development, RAG systems, data science, bioinformatics, machine learning, and performance optimization. Experienced Claude Code developer who has shipped autonomous coding agents and published a deep-dive series on building them.
- Research & Technical Writing: Published researcher who knows how to analyze and interpret data, validate findings, and explain results clearly.
- Full-Stack Development: Comfortable designing relational and Neo4j graph database schemas, building FastAPI backends, interacting with external APIs, and creating responsive frontends.
- Leadership Experience: Project management experience combined with teaching experience makes me effective at communicating information and facilitating idea generation.