About Me
Hi, I’m Tirath — a final-year Computer Engineering student from Mumbai focused on AI systems, machine learning research, and building practical tools around LLMs and multimodal AI.
Most of my work sits at the intersection of research and engineering: from multilingual hate speech detection and peer-review transparency systems to code transpilation and agentic AI frameworks. I enjoy turning messy, ambiguous research ideas into systems that can actually be deployed, evaluated, and used by people.
Outside of academics, I’m someone who genuinely enjoys deep technical rabbit holes, mentoring students, exploring cities through food, and having long conversations about technology, academia, and where AI is headed next.
Currently Exploring
- Mechanistic interpretability and reasoning traces in LLMs
- Multi-agent systems for scientific workflows
- Small language model finetuning
- Retrieval + argumentation-based reasoning systems
- AI systems that improve research peer review and evaluation
Right Now
- Finishing my B.Tech in Computer Engineering
- Preparing to begin an MSCS journey in the US
- Working on research around AI-assisted peer review and multimodal systems
- Experimenting with SLM finetuning and agentic workflows
What Drives Me
What excites me most about AI is not just model capability, but the possibility of building systems that augment human thinking — helping people reason better, review information more carefully, and navigate complex decisions more transparently.
"I’m especially drawn toward problems where engineering, research, and human judgment intersect."