# Mohammed Farhan Baluch

> AI Engineer, Toronto, Canada. I build multi-agent systems, and the evaluation that proves they're safe to ship.

AI Engineer at ML Arteka (formerly mobileLIVE), an AI and digital consultancy. I design multi-agent LLM systems for enterprise clients in telecom and banking, then build the evaluation, test automation and rollout that make them trustworthy. Before that, an MSc in AI with five peer-reviewed publications.

Focus: Multi-agent systems; Agent evaluation; AI quality engineering; Legacy modernization with agents. Open to: Senior AI engineering, AI solutions architecture.

## Recent work log

- 2026-10-19: **Extended coverage reporting to mobile test suites.** Mobile teams write tests in Java with TestNG, which the first parser couldn't read. Added a Java-aware parser so their coverage shows up in the same weekly report. (https://farhanbaluch.com/work/coverage-insights.md)
- 2026-10-14: **Published a structure and naming standard for 20+ test repositories.** Every team laid out its repository differently, which made automated reporting brittle. Wrote one standard for folders, tags and naming, with a migration checklist per team.
- 2026-10-08: **Pitched an agentic quality-intelligence offering.** Turned the risk-based test design results into a service offering, with a demo, a pilot plan and the evidence from the bug-injection study. (https://farhanbaluch.com/work/risk-based-test-design.md)
- 2026-10-07: **Demoed multi-agent documentation for legacy firmware.** An orchestrator splits the codebase across explorer agents, and an independent reviewer agent rejects any claim it can't trace to a cited line. (https://farhanbaluch.com/work/legacy-docs-agents.md)
- 2026-10-06: **Shipped the first weekly automation-coverage report.** 21 teams, measured straight from code instead of spreadsheets. The first run put regression automation coverage at 81.7%. (https://farhanbaluch.com/work/coverage-insights.md)
- 2026-09-29: **Started coverage measurement from source.** Leadership's automation numbers came from hand-kept spreadsheets nobody could trace. Prototyped a zero-dependency pipeline that clones every team repository and counts tests by tag. (https://farhanbaluch.com/work/coverage-insights.md)
- 2026-09-22: **Launched a searchable catalog of 27 agents.** Mapped every agent and skill to one of seven SDLC phases and published them in a docs portal, so engineers find an agent before they write a new one.
- 2026-09-15: **Added a reviewer agent to the documentation pipeline.** Early drafts read well but overstated what the code did. A separate reviewer agent now checks each section against the source and sends unsupported claims back. (https://farhanbaluch.com/work/legacy-docs-agents.md)

Full log: https://farhanbaluch.com/log.md

## Projects

- [Automation coverage, measured from source](https://farhanbaluch.com/work/coverage-insights.md): 81.7%, regression automation coverage across 21 teams, now measured weekly
- [Agent and skill distribution platform](https://farhanbaluch.com/work/agent-distribution.md): 18 repos, on versioned agent packages, from 7 packages
- [Agent evaluation pipeline](https://farhanbaluch.com/work/agent-evaluation.md): 10+, quality dimensions scored per agent, from traces as well as answers
- [Multi-agent legacy documentation](https://farhanbaluch.com/work/legacy-docs-agents.md): 16 sections, every claim verified against cited lines
- [Risk-based test design with agents](https://farhanbaluch.com/work/risk-based-test-design.md): 9 of 9, injected bugs caught, against 1 of 9 for the standard set
- [Market research agents for retail banking](https://farhanbaluch.com/work/market-research-agents.md): 99.3%, extraction accuracy across 20+ bank websites
- [Estimation platform for an enterprise bank](https://farhanbaluch.com/work/estimation-platform.md): 80%, less manual estimation effort
- [Agentic legacy modernization and code intelligence](https://farhanbaluch.com/work/legacy-modernization.md): 70%, shorter modernization timeline
- [Autonomous testing for a conversational voice agent](https://farhanbaluch.com/work/voice-agent-testing.md): <10 min, to generate branch-covering test cases for a call flow

## Skills by layer

- Interfaces: Realtime voice agents, LiveKit, OpenAI Realtime, Next.js, React, FastAPI
- Orchestration: Multi-agent systems, LangGraph, Google ADK, Claude Agent SDK, Claude API, Model Context Protocol, GitHub Copilot agents, Agent skills
- Models: Claude, Gemini, OpenAI, Llama 3.2 with QLoRA
- Knowledge: GraphRAG, Neo4j, Azure AI Search, Databricks, Crawl4AI, PostgreSQL
- Evaluation and quality: LLM-as-judge, DeepEval, Promptfoo, OpenTelemetry, Playwright, Pact, TestNG, Risk-based test design, Coverage metrics
- Delivery: Spec-driven development, Agent packaging, Legacy modernization, Java 21 and Spring Boot, JasperReports, MLflow
- Cloud and languages: Azure, AWS, Python, TypeScript, Java

## Résumé

- Text: https://farhanbaluch.com/resume.md
- PDF: https://farhanbaluch.com/resume.pdf

## Contact

- Email: farhanbaluch1301@gmail.com
- LinkedIn: https://www.linkedin.com/in/farhan1301
- GitHub: https://github.com/farhan1301