Looks like you have JavaScript disabled. For the full Twine experience, you will need to re-enable it.

I am a Senior AI/ML Engineer with nearly a decade of experience across Image Sensor development and ADAS. I specialize in deploying high-impact solutions, from building Agentic AI and RAG-based applications using Langchain to optimizing hardware design workflows with Multi-Objective Optimization. My work at Sony Semiconductor Solutions notably reduced simulation times from 12 hours to 2 hours through AI integration. I pair deep technical expertise in Python, TensorFlow, and Computer Vision. Based in Japan, I am a collaborative leader proficient in both English and Japanese.…I am a Senior AI/ML Engineer with nearly a decade of experience across Image Sensor development and ADAS. I specialize in deploying high-impact solutions, from building Agentic AI and RAG-based applications using Langchain to optimizing hardware design workflows with Multi-Objective Optimization. My work at Sony Semiconductor Solutions notably reduced simulation times from 12 hours to 2 hours through AI integration. I pair deep technical expertise in Python, TensorFlow, and Computer Vision. Based in Japan, I am a collaborative leader proficient in both English and Japanese.

Varun Vivek Pai

AI Engineer, Data Scientist, AI Prompt Engineer





I am a Senior AI/ML Engineer with nearly a decade of experience across Image Sensor development and ADAS. I specialize in deploying high-impact solutions, from building Agentic AI and RAG-based applications using Langchain to optimizing hardware design workflows with Multi-Objective Optimization. My work at Sony Semiconductor Solutions notably reduced simulation times from 12 hours to 2 hours through AI integration. I pair deep technical expertise in Python, TensorFlow, and Computer Vision. Based in Japan, I am a collaborative leader proficient in both English and Japanese.…I am a Senior AI/ML Engineer with nearly a decade of experience across Image Sensor development and ADAS. I specialize in deploying high-impact solutions, from building Agentic AI and RAG-based applications using Langchain to optimizing hardware design workflows with Multi-Objective Optimization. My work at Sony Semiconductor Solutions notably reduced simulation times from 12 hours to 2 hours through AI integration. I pair deep technical expertise in Python, TensorFlow, and Computer Vision. Based in Japan, I am a collaborative leader proficient in both English and Japanese.

Available to hire

I am a Senior AI/ML Engineer with nearly a decade of experience across Image Sensor development and ADAS. I specialize in deploying high-impact solutions, from building Agentic AI and RAG-based applications using Langchain to optimizing hardware design workflows with Multi-Objective Optimization. My work at Sony Semiconductor Solutions notably reduced simulation times from 12 hours to 2 hours through AI integration. I pair deep technical expertise in Python, TensorFlow, and Computer Vision. Based in Japan, I am a collaborative leader proficient in both English and Japanese.

Skills

Experience Level

Expert

Expert

Intermediate

Intermediate

Intermediate

Beginner

Beginner

Beginner

Language

English

Advanced

Japanese

Advanced

Work Experience

Senior Lead Engineer at Sony Semiconductor Solutions (via Quest Global Japan)

April 1, 2018 - Present

Python-based Streamlit web application design and development using LLM, Prompt Engineering and retrieval augmented generation (RAG) for context-based generation; vector databases; AI tooling with LangChain and FastAPI for data-driven image sensor design; AI model development using MOO, neural networks and surrogate modeling to accelerate design workflows; trained and validated AI models for pixel design integrated with TCAD simulations on IBM LSFF cluster; improved performance by reducing simulation time from 12 hours to 2 hours; UI and visualization application development for data analysis.

Senior Software Engineer at Quest Global

September 1, 2014 - March 1, 2018

Experience using computer vision and deep learning frameworks for ADAS projects like traffic sign recognition and pedestrian detection. Achieved 90% accuracy on real-time dash cam videos. Model training and validation using Caffe, NVIDIA and preparing image datasets and data annotations for NN training. Developed a C# .NET WPF-based plug-in framework to build a graph editor analogous to Simulink; deployed MATLAB-based optimization algorithms via .NET/C# libraries. 360-degree surround-view development using CCD cameras with real-time implementation on Linux OS for NVIDIA Tegra K1 platform.