Summary
Lakshya Gupta is an applied ML research scientist with five years of experience building real-world AI systems across video codecs, LLM reasoning, world models and reinforcement learning. Currently at Skyfall AI, he focuses on safe long-horizon planning for complex enterprise systems, having previously shipped a novel Gaussian-splatting neural video codec at AMD that delivered real-time decoding and large bandwidth savings. He combines strong academic training (MS Applied Computing, Univ. of Toronto; BTech in CS) with hands-on engineering across Python/PyTorch, CUDA/C++ and ONNX to move research into production. An open-source enthusiast and former GSoC contributor, he also brings teaching experience translating complex concepts to diverse audiences. He’s notable for blending explainable, manipulable AI architectures with pragmatic optimizations that prioritize integration with existing pipelines.
5 years of coding experience
1 year of employment as a software developer
Bachelor of Technology - BTech, Computer Science, 9.14, Bachelor of Technology - BTech, Computer Science, 9.14 at Vellore Institute of Technology
Master of Science - MS, Applied Computing - AI Concentration, 3.92, Master of Science - MS, Applied Computing - AI Concentration, 3.92 at University of Toronto
ITL Public School