About

I am Jaya Sathiya Ravichandran, a B.Tech in Biomedical Engineering graduate from PSG College of Technology, class of 2026. During my studies, I worked on computer vision and AI/ML projects, including building a deep learning model for diabetic foot ulcer risk detection and an AI-driven phototherapy optimization framework. I also gained experience as an Algorithm Developer Intern at HTIC, IITM, where I developed optical tracking pipelines, and as a TPRM Analyst Intern at Cyfotok, where I assessed vendor security risks. My strongest skills include C++, Python, MATLAB, OpenCV, TensorFlow, and tools like 3D Slicer and Unity. I am now looking for my first full-time role as a Software Engineer.

Experience

1. Algorithm Developer - Intern

   Computer Vision-HTIC, IITM

   Feb 2026 – Sep 2026

   Built an optical tracking pipeline in C++/OpenCV using PnP-based pose estimation, with MATLAB camera calibration. Implemented pivot calibration, camera-to-CT rigid registration and automated ASTM divot-point data collection. Streamed real-time tool poses to 3D Slicer via a Python OpenIGTLink server. Evaluated tracking accuracy (RMS error, standard deviation) across the working volume and compared cameras. Built multi-face marker tracking for a monocular and an AR headset (Unity).

2. TPRM Analyst - Intern

   Cyfotok

   Sep 2025 – Nov 2025 · Coimbatore, India

   Performed third-party risk assessments of vendor security posture, inherent risk and control effectiveness. Reviewed SOC 2 reports, audit documentation and evidence to identify control gaps. Supported vendor onboarding through due diligence and security questionnaire evaluation for compliance. Documented risk findings in Archer and updated the SIG.

Education

1. PSG College of Technology

   BE (Biomedical Engineering)

   2022 – 2026 · Grade: 8.43 (cgpa)

2. SRV Girls Higher Secondary School

   12th Standard

   2021 – 2021 · Grade: 94.7 %

Projects

• Multimodal Thermographic Analysis for Angiosome-Based Early Detection of DFU Risk

  • Python
  • TensorFlow
  • OpenCV

  Built DFU-ThermoNet (Spiking Transformer) for diabetic foot ulcer risk classification, reaching 96.8% accuracy on 1,670 thermal images from 167 subjects. Mapped risk across 4 angiosome regions with confidence scores and an auto-generated report in a web interface.

• AI-Driven Optimization Framework for Phototherapy in Skin Disease

  • Python
  • TensorFlow
  • OpenCV

  Built a 12-class skin disease classifier hybrid model with 95% accuracy on 19,559 images, after merging 23 overlapping classes and balancing data with SMOTE. Added Grad-CAM explainability and PASI/VASI/SCORAD severity grading with a web interface.

Certifications

• Google Professional Cybersecurity Certificate

  Google

• ISO27701 Lead Auditor

  ISO27701 Lead Auditor-Mastermind

• Generative AI in Healthcare

  Microsoft

• TryHackMe SOC L1

  TryHackMe

• Data Analyst, Networking

  CISCO

Achievements

• Multimodal Thermographic Analysis for Angiosome-Based Early Detection of DFU Risk

  2025

  Built DFU-ThermoNet (Spiking Transformer) for diabetic foot ulcer risk classification, reaching 96.8% accuracy on 1,670 thermal images from 167 subjects. Mapped risk across 4 angiosome regions with confidence scores and an auto-generated report in a web interface.

• AI-Driven Optimization Framework for Phototherapy in Skin Disease

  2025

  Built a 12-class skin disease classifier hybrid model with 95% accuracy on 19,559 images, after merging 23 overlapping classes and balancing data with SMOTE. Added Grad-CAM explainability and PASI/VASI/SCORAD severity grading with a web interface.