About Me
I am a systems-oriented engineer with strong foundations in machine learning, robotics, and software architecture. My work spans training deep learning models, integrating perception with control systems, and deploying AI pipelines that operate reliably in real environments. I value clarity in design, robustness over novelty, and understanding failure modes before scaling solutions.
Experience
Technical Projects
Intelligent Document Agent
IDP-v2 is a sophisticated, production-grade Intelligent Document Processing platform. It is not just a prototype but a well-architected system designed for scalability, observability, and flexibility. The integration with RAG-Anything (as a vendored submodule) provides state-of-the-art multimodal parsing capabilities (MinerU/Docling), while the core platform handles the orchestration, state management, and API layer.
GPT-Style Transformer
Implemented a GPT-style Transformer architecture from first principles, including tokenization, positional embeddings, and multi-head self-attention.
Autonomous Multi-Agent
Built an orchestration layer for autonomous agents using Model Context Protocol, enabling LLMs to interact with external tools and APIs.
Real-Time Basketball Detection
Developed an object detection pipeline using custom and synthetic datasets to improve robustness and consistency during real-time inference.
Autonomous Line Follower
Designed and built an embedded robotic system with sensor-driven control logic and smooth trajectory correction on dynamic paths.
Technical Arsenal
Deep Learning & AI
Robotics & Embedded
Languages
Cloud & Infra
Frameworks
Certifications
Deep Learning Specialization
DeepLearning.AI & Andrew Ng
Mastered Neural Networks, CNNs, Sequence Models, and Hyperparameter Tuning. Built and trained models for object detection and NLP tasks from scratch.
Machine Learning Specialization
Stanford University & DeepLearning.AI
Comprehensive training in Supervised (Regression/Classification) and Unsupervised Learning. Implemented recommender systems and reinforcement learning agents.
Data Structures & Algorithms in C++
Coding Ninjas
Developed a rigorous understanding of data structures, algorithm design, and low-level programming concepts in C++. Emphasized problem decomposition, trade-off analysis, and performance optimization.
Education
Sardar Vallabhbhai National Institute of Technology (NIT)
B.Tech in Mechanical Engineering
Indian Institute of Technology, Madras (IITM)
B.S. in Electronic Systems