Madhusudhan Chandar

About Madhusudhan Chandar

I’m an AI/ML enthusiast with strong hands-on experience in computer vision, deep learning, and agentic AI systems. I’ve worked on real-time solutions such as parking-lot speed detection and ADAS object detection, focusing on model accuracy, tracking, and end-to-end AI pipelines. I also build agentic AI applications that enable autonomous reasoning, decision-making, and intelligent workflows. Passionate about creating practical, scalable AI solutions that deliver real-world impact.

Experience

1. Project Intern

   Oracle

   Jan 2026 — Present

   Architected an End-to-End LLM Security Model: Engineered a custom \"Prompt Firewall\" to classify and validate inputs. Owned the complete ML lifecycle, from curating and preprocessing custom training datasets to fine-tuning a DeBERTa architecture for high-precision text classification.• Engineered an Autonomous Data-Extraction Agent: Designed an intelligent agent capable of parsing complex, unstructured API documentation and autonomously generating structured, production-ready JSON schemas.• Built Advanced RAG & Ingestion Pipelines: Developed a highly accurate hierarchical retrieval system. Transformed raw HTML and PDF sources into searchable data using custom page-indexing and tree-based search algorithms.

Education

• CMR Institute Of Technology

  Bachelor of Engineering - BE, Computer Science Engineering (Artificial Intelligence and Machine Learning)

  2022 — 2026

• Vydehi School of Excellence

  Higher Secondary CBSE, Science