I am based in Delhi, Delhi, India. My headline is: School of Engineering | JNU | Data Intern @ SherlockAI | AI Research & Engineering Enthusiast | AI Agents • Reinforcement Learning • Self-Supervised & Multimodal AI.
I am a Data intern at SherlockAI by Infinite Analytics (Mar 2026 to Present), Delhi, India.
I was an AI/ML intern at Clikkin (Jul 2025 to Oct 2025), Delhi, India.
My education is Jawaharlal Nehru University: Bachelor of Technology - BTech, Electrical, Electronics and Communications Engineering (2023).
My skills include Kotlin, C++, Problem Solving, Java, C (Programming Language), Android Development, Python (Programming Language), GitHub, and Git. My interests include AI research and engineering, AI agents, reinforcement learning, self-supervised AI, and multimodal AI.
I earned the certification Problem solving, Apr 2024, HackerRank.
My publication is “Adaptive Depth Transformer-DQN: Dynamic Computation for Value-Based Reinforcement Learning,” published on Techrxiv on Jan 27, 2026. It proposes dynamically adjusting Transformer computation depth per state using intermediate Q-value heads and uncertainty-aware halting, aiming to reduce unnecessary computation while preserving Bellman consistency. The abstract reports performance comparable to fixed-depth Transformer-DQN while significantly reducing average inference depth, without giving a specific numerical result. My work on this topic is described in problem solving and adaptive computation.