About
Dhinakaran Thangaraj
AI/ML Developer · AI Security · Agentic Systems
I am an AI/ML developer based in Chennai, India, currently working at SQ1 Cybersecurity Solutions. My work sits at the intersection of applied AI, security engineering, and governance, risk and compliance (GRC). I build systems that collect and validate evidence, reason over controls, assess risk, and support security teams with more reliable workflows.
My interests include LLM engineering, multi-agent systems, retrieval-augmented generation, AI security, and practical machine learning infrastructure. I also build open-source developer tools that make AI systems easier to use and maintain.
Current focus
- Agentic GRC: AI-assisted security evidence processing, control assessment, vendor risk, and governed knowledge retrieval.
- LLM systems: LangGraph agents, RAG pipelines, provider routing, and production Python APIs.
- AI security research: Prompt injection, cross-tool boundaries, and security risks in multi-skill agent systems.
- Developer tooling: Open-source Python packages for model tool calling, ML environment diagnostics, and backend scaffolding.
Selected work
- toolproxy — tool calling for LLMs with and without native function-call support.
- mlenvdoctor — diagnose and repair machine-learning environments.
- boostapi — FastAPI project scaffolding.
- AI-Based Mock Interview Platform — peer-reviewed publication, April 2025.
You can explore my experience, projects, publications, and open-source work.