Sampurna Roy profile photo 1 on Tal

Sampurna Roy

AI/ML Developer (Python and AI automation Engineer) Selec Controls Pvt. Ltd.

Mumbai

< 1 year of work experience

Selec Controls Pvt. Ltd.
Jul 20264 mosPresent

Navi Mumbai

Building and deploying production AI systems for engineering, manufacturing, and field service teams. • Designed and built SELTALKS, a multimodal RAG assistant (Gemini + LangChain + FAISS) that helps engineers and customers diagnose issues across Selec's product line using text, code, or images — with an automated OCR pipeline that ingests wiring diagrams and tables straight from technical PDFs. • Built an LLM-as-a-judge evaluation layer that scores every AI response against retrieved context in real time, plus a self-learning feedback loop that lets verified corrections be injected into the live system without redeployment. • Built a debug copilot for our Grid Tie Inverter production line, combining structured SOP matching with semantic search, and a text-to-SQL assistant that lets operators query production/defect data in plain English. • Developed an AI-based computer vision system for automated PCB wiring and solder inspection, using OpenCV (HSV color masking, contour detection, spatial analysis) to verify wire routing and origins, and a YOLOv8 model to detect and count solder joints/valleys across PCB images captured at multiple angles. Combined both into an automated Pass/Fail evaluation pipeline with a Tkinter GUI for visual fault indication — replacing manual visual inspection with a repeatable, automated check on the production line. • Working across the full stack — model integration, retrieval, evaluation, reliability, and security — to get these systems trusted enough for daily production use.
ROBOSERV 4i
Feb 20265 mosJun 2026

Navi Mumbai

Worked as a Software, Python & AI Developer Intern, developing AI-powered automation solutions for the DuctBot industrial robotic inspection system. Built and integrated Computer Vision pipelines using YOLO and OpenCV for real-time defect detection, image processing, and visualization of AI predictions on live video streams. Developed Python applications using Kivy for robot control, inspection monitoring, and camera management. Designed and managed databases using SQLite and Amazon RDS to store inspection data, AI outputs, and system logs. Optimized AI inference and application performance on Raspberry Pi through multi-threading and implemented secure data handling mechanisms. Collaborated on developing intelligent automation solutions by integrating Machine Learning, Computer Vision, and Python-based backend components.
Monster India
Jun 20253 mosAug 2025

Noida

Worked as a Software Engineer Intern, contributing to the development of a full-stack e-commerce application using Java, Spring Boot, HTML, CSS, JavaScript, and MySQL. Designed and implemented backend APIs, user authentication, shopping cart, and product management modules. Developed and integrated an AI-powered chatbot using Google Gemini that provided personalized product recommendations based on user events and occasions, analyzed customers' purchase history to generate shopping insights, and suggested products tailored to individual buying patterns. Built an interactive analytics dashboard to visualize user spending across different product categories, enabling users to track expenses and make informed purchasing decisions. Collaborated on backend development, API integration, database management, and AI-driven feature implementation to enhance user experience.

Sampurna’s AI Stack

antigravity
antigravity
claude code
claude code
claude
claude
cursor
cursor
gemini
gemini
github copilot
github copilot

Education

Indian Institute of Technology, Patna

Master of Technology, Artificial Intelligence and Data Science Engineering

Indian Institute of Technology, Patna

20262 yrs2028
amity University

Bachelor of Technology - BTech, Computer Science

amity University

20224 yrs2026

Join Sampurna and 10k+ other professionals on

Tal Boss

Tal is the app where bosses meet engineers & builders directly, without middlemen

Download Tal Boss

Sampurna Roy on Tal

Headline

AI/ML Developer (Python and AI automation Engineer) Selec Controls Pvt. Ltd.

About

AI/ML Engineer, Data Scientist, and Python Developer with hands-on experience building and deploying production-grade AI systems, data-driven solutions, and full-stack applications. Strong foundation in Artificial Intelligence, Machine Learning, Data Science, Computer Vision, NLP, and software development, with experience taking projects from research and prototyping to real-world deployment. IBM certified, with 3 published research papers and a Best Paper Award for an AI research framework. Combines strong programming and problem-solving skills with practical experience in developing reliable, scalable, and intelligent solutions.

Location

Mumbai

AI stack

antigravity

claude code

claude

cursor

gemini

github copilot

Work experience

AI/ML Developer (Python and AI automation Engineer) at Selec Controls Pvt. Ltd., Jul 2026 - Present · 4 mos, Navi Mumbai, Building and deploying production AI systems for engineering, manufacturing, and field service teams. • Designed and built SELTALKS, a multimodal RAG assistant (Gemini + LangChain + FAISS) that helps engineers and customers diagnose issues across Selec's product line using text, code, or images — with an automated OCR pipeline that ingests wiring diagrams and tables straight from technical PDFs. • Built an LLM-as-a-judge evaluation layer that scores every AI response against retrieved context in real time, plus a self-learning feedback loop that lets verified corrections be injected into the live system without redeployment. • Built a debug copilot for our Grid Tie Inverter production line, combining structured SOP matching with semantic search, and a text-to-SQL assistant that lets operators query production/defect data in plain English. • Developed an AI-based computer vision system for automated PCB wiring and solder inspection, using OpenCV (HSV color masking, contour detection, spatial analysis) to verify wire routing and origins, and a YOLOv8 model to detect and count solder joints/valleys across PCB images captured at multiple angles. Combined both into an automated Pass/Fail evaluation pipeline with a Tkinter GUI for visual fault indication — replacing manual visual inspection with a repeatable, automated check on the production line. • Working across the full stack — model integration, retrieval, evaluation, reliability, and security — to get these systems trusted enough for daily production use.

Python Developer (AI/ML) at ROBOSERV 4i, Feb 2026 - Jun 2026 · 5 mos, Navi Mumbai, Worked as a Software, Python & AI Developer Intern, developing AI-powered automation solutions for the DuctBot industrial robotic inspection system. Built and integrated Computer Vision pipelines using YOLO and OpenCV for real-time defect detection, image processing, and visualization of AI predictions on live video streams. Developed Python applications using Kivy for robot control, inspection monitoring, and camera management. Designed and managed databases using SQLite and Amazon RDS to store inspection data, AI outputs, and system logs. Optimized AI inference and application performance on Raspberry Pi through multi-threading and implemented secure data handling mechanisms. Collaborated on developing intelligent automation solutions by integrating Machine Learning, Computer Vision, and Python-based backend components.

Technology (Software Developer) at Monster India, Jun 2025 - Aug 2025 · 3 mos, Noida, Worked as a Software Engineer Intern, contributing to the development of a full-stack e-commerce application using Java, Spring Boot, HTML, CSS, JavaScript, and MySQL. Designed and implemented backend APIs, user authentication, shopping cart, and product management modules. Developed and integrated an AI-powered chatbot using Google Gemini that provided personalized product recommendations based on user events and occasions, analyzed customers' purchase history to generate shopping insights, and suggested products tailored to individual buying patterns. Built an interactive analytics dashboard to visualize user spending across different product categories, enabling users to track expenses and make informed purchasing decisions. Collaborated on backend development, API integration, database management, and AI-driven feature implementation to enhance user experience.

Education

Master of Technology, Artificial Intelligence and Data Science Engineering, Indian Institute of Technology, Patna

Bachelor of Technology - BTech, Computer Science, amity University

Public links

github: https://github.com/sampurna1524

sampurna1524.github.io: https://sampurna1524.github.io/