Kiran Shahi is a Software Developer, AI Engineer, and Computer Vision Researcher passionate about developing intelligent, real-time, and visually rich applications.
He currently works at MBS Survey Software Limited, where he contributes to building advanced solutions that integrate Computer Vision, 3D Graphics, and Artificial Intelligence to enhance industry workflows and visual analysis.
His work bridges the gap between academic research and practical engineering, focusing on transforming complex algorithms into efficient, production-ready software.
- π» Skilled in C#, Python, and TypeScript for cross-platform and high-performance applications
- π§© Specialized in Computer Vision, AI-driven Graphics, and Deep Learning Architectures
- βοΈ Experienced in 3D visualization, image processing, and real-time rendering
- π Interested in Edge AI, video matting, and explainable AI for transparent decision-making
- π± Constantly learning and experimenting with emerging technologies
Core Expertise:
- Languages: C#, Python, TypeScript/JavaScript
- Frameworks & Libraries: .NET 8, PyTorch, TensorFlow, OpenCV, Angular 20
- Domains: Computer Vision, Machine Learning, 3D Graphics, Edge AI
- Tools: Git, Visual Studio, VS Code, SQL Server, RhinoCommon SDK
Kiran developed a high-performance AI-powered video background matting system that enhances online conferencing and live streaming experiences.
Highlights:
- β‘ Real-time background replacement with CNNs and ConvLSTM
- π§ Maintains temporal consistency and natural visual quality
- π Privacy-first approach β processes data locally
- π§© Built using Python, OpenCV, and Deep Learning frameworks
Keywords: computer vision, background matting, deep learning, AI, real-time video processing, privacy-preserving inference.
- π§ͺ Active contributor to open-source AI research and visual computing projects
- π Published works in medical imaging and explainable AI domains
- π Building innovative SaaS tools for glint and glare simulation, solar analysis, and CAD integration
- π MSc in Artificial Intelligence from Brunel University London