Yash Masane is a data scientist with eight years of experience building AI systems that perform reliably in production, with a particular strength in signal estimation and time-series modeling. He designed a hybrid CNN-LSTM attention model for Direction of Arrival estimation that outperformed traditional methods, achieving R² of 0.95 and MAE of 2.79°, and applied feature selection and metaheuristic optimization to cut validation loss further. Comfortable across the ML lifecycle, he prioritizes testable, production-ready code—summed up by his GitHub credo, "if it ain't tested, it's broken." Currently at Cogninest AI, he pairs a strong mathematical background (MSc in Maths and Scientific Computing) with hands-on engineering to turn research-grade models into deployable systems. He also has experience tutoring mathematics and shipping end-to-end pipelines using TensorFlow, Keras, and scikit-learn. Less obvious: he blends classical signal-processing insight with modern attention mechanisms and optimization techniques to make models both interpretable and robust.
8 years of coding experience
M.Sc Maths and Scientific Computing, M.Sc Maths and Scientific Computing at Motilal Nehru National Institute Of Technology
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.