Hi, I'm

Dheepak Shakthi

CSE Student | AI, Machine Learning & Full Stack Developer

Dheepak Shakthi

About Me

B.Tech Computer Science Engineering student specializing in AI and Data Analytics with hands-on experience in Machine Learning and full-stack application development, with a focus on enterprise-scale projects and innovative software engineering solutions.

Programming Languages

Python, Java, C, C++, C#, JavaScript, SQL

AI & ML

PyTorch, TensorFlow & TensorBoard, OpenCV, Hugging Face, Ollama & OpenAI API models

App Development

Node.js, Express, React.js, Next.js, React Native, Flask, Django, Flutter, .NET, REST APIs

Database Systems

MySQL Workbench, Supabase (BaaS) with PostgreSQL, MongoDB Atlas

DevOps, Tools & Software

Docker, Git, CI/CD pipelines, Tableau, Power BI, Vim, Linux

Languages

Tamil — Native  |  English — Proficient  |  Hindi — Limited working proficiency

Other Skills

Unity Game Engine, Blender, Adobe Premiere pro, Audacity

Internships

AI Research Intern

HTIC, IIT-Madras Research Park

May 2025 – June 2025  ·  Chennai, Tamil Nadu

Worked in medical imaging for detection and classification of abnormalities and anatomical landmarks in real-time endoscopic video inference on edge devices. Developed knowledge distillation pipelines, optimized deep learning models, benchmarked all models, and documented their performance.

AI Intern

INEM Technologies

December 2025 – January 2026  ·  Remote

Implemented database and designed pipeline for implementing AI features in an AI-assisted Ed-Tech web application platform developed using Django.

Featured Projects


Publications

  • IEEE CICT 2025 — Published Paper
    "Penile Curvature Quantification in Hypospadias Surgery using Deep Learning and Circle-Fit Optimization" — presented and published at IEEE CICT 2025, IIITDM Kancheepuram.
    DOI: 10.1109/CICT67193.2025.11398889
  • Journal of Pediatric Urology — Published Work
    "CurvAssist: an AI assisted pipeline for penile shaft segmentation/curvature measurement in children with hypospadias"
    DOI: 10.1016/j.jpurol.2026.105877

Get in Touch