Tanmay Garg

Tanmay Garg

ML Engineer @ Adobe | M.S. ML @ CMU | Agentic AI, LLM Evaluation | Physics-informed diffusion, Scientific ML, Explainable AI.

ML Engineer and Researcher

I’m Tanmay Garg, an ML Engineer at Adobe and M.S. in Machine Learning graduate from Carnegie Mellon University. I build agentic AI systems and the evaluation infrastructure that tells you whether they work, and I study how generative models can respect physics.

At Adobe, I work on the Experience Intelligence team building LLM evaluation infrastructure and agentic AI systems for Adobe Firefly. My work spans LLM evaluation frameworks, multi-agent orchestration with MCP and LangGraph, and prompt engineering for enterprise AI applications.

I completed my M.S. at CMU working as a Graduate Research Assistant with Prof. Aarti Singh and Prof. Barnabas Poczos on physics-informed diffusion models for cosmology and multi-modal PDEs, bridging generative AI with physics-based learning. This work was published at the AI and PDE Workshop, ICLR 2026.

Prior to CMU, I completed my undergraduate studies in Computer Science and Engineering at the Indian Institute of Technology, Hyderabad, where I was associated with the Machine Learning and Vision Group under Prof. Vineeth Balasubramanian, leading to publications at ICCV 2023 and AAAI 2024.

Read about my research on the Research page and Google Scholar.

Research Interests:

  • Agentic AI & Multi-Agent Systems
  • LLM Evaluation
  • Diffusion Models
  • Physics-Informed Machine Learning
  • Explainable AI
  • Computer Vision & Transfer Learning

Personal Interests:

Outside work I play guitar, mostly rock, blues and acoustic, and compete in squash. I also read about astronomy and the fundamentals of physics, and I try to keep up with new technology.