Dibakar Saha
Senior Data Scientist
West Bengal, India
Summary
👤
Senior🎓
Top SchoolDibakar Saha is a Senior Data Scientist with 9 years of experience building scalable AI systems that translate complex healthcare data into clinically meaningful insights. With a BTech in Computer Engineering, he progressed from data engineering to leading advanced AI applications in medical imaging at nference and now at a stealth AI startup. He combines hands-on model development and production engineering to deliver solutions that improve patient outcomes, including end-to-end pipelines for imaging and structured data. A lifelong programmer, Dibakar also contributes practical ML projects—such as a real-time sign-language CNN integrating TensorFlow, OpenCV, and an SQLite-backed interface—demonstrating his full-stack ML engineering chops. Based in West Bengal, India, he blends domain fluency in life sciences with pragmatic software design to move prototypes into reliable, deployable systems. He’s driven by a curiosity for turning messy biomedical data into reproducible, impactful tools that accelerate medical breakthroughs.
9 years of coding experience
7 years of employment as a software developer
Bachelor of Technology - BTech, Computer Engineering, Bachelor of Technology - BTech, Computer Engineering at Bengal College of Engineering and Technology 125ISC, Computer Science, ISC, Computer Science at Holy Child School, Jalpaiguri
English, Bengali, Hindi
Github Skills (10)
mask-rcnn10
keras10
computer-vision10
faster-rcnn10
machine-learning10
opencv10
tensorflow10
python10
image-processing10
sqlite8
Programming languages (3)
Github contributions (5)
EvilPort2/Sign-Language
Feb 2018 - Aug 2019
A very simple CNN project.
Role in this project:
ML Engineer
ML EngineerContributions:1 release, 59 commits, 3 PRs in 1 year 5 months
Contributions summary:Dibakar primarily contributed to the development and refinement of a CNN project for sign language recognition. Their work involved implementing the CNN model using TensorFlow and Keras, training the model, and integrating it with OpenCV for real-time gesture recognition from a webcam feed. The contributions include adding and modifying the necessary code for the CNN model, data processing, and creating a user interface to visualize the recognition results. The user also incorporated an SQLite database for mapping recognized gestures to text.
pythondeep-learningmachine-learningneural-networktensorflow
EvilPort2/SimpleGestureRecognition
Aug 2017 - Jun 2018
Contributions:1 release, 28 commits, 28 pushes in 10 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.