Som Tambe
Milan, Lombardy, Italy
Summary
👤
Senior🎓
Top SchoolSom Tambe is a Founding Industrial Design Engineer with nine years’ experience turning complex ideas into rugged, manufacturable products across defense, aerospace, oil & gas, and consumer markets. Trained at IIT Kanpur and Politecnico di Milano, he leads end-to-end mechanical design—from concept sketches and CNC-optimized housings to modular actuation, automated test rigs, and firmware integration using Python/C/C++. At Armatrix he drove the architecture of military-grade, high-DOF robotic arms and built remote automated testing pipelines using Raspberry Pi/ESP32, demonstrating a rare blend of hands-on prototyping and production-grade supply chain management. He’s also an active open-source contributor to the Julia ecosystem, adding orthogonal initialization to Flux.jl and improving interpolation test coverage, signaling strong ML and numerical computing chops beyond mechanical engineering. Based in Milan, Som thrives at the intersection of industrial design, embedded systems, and computer vision, with a focus on serviceability, robustness, and elegant integration of software and hardware.
9 years of coding experience
Indian Institute of Technology KanpurMaster of Science - MS, Design & Engineering - Progetto e Ingegnerizzazione del Prodotto Industriale, Master of Science - MS, Design & Engineering - Progetto e Ingegnerizzazione del Prodotto Industriale at Politecnico di MilanoHigh School Diploma, Science, High School Diploma, Science at Shri TP Bhatia College of ScienceGrade 10, Science, Grade 10, Science at Children's Academy Group of Schools
English, Hindi, Marathi, Gujarati, Italian
Stackoverflow
21reputation
2kreached
0answers
3questions
Github Skills (18)
testing10
machine-learning10
interpolation10
gradient10
deep-learning10
flux10
fluxor10
julia10
linear-algebra9
unit-testing8
neural-network8
data-science8
transfer-learning6
try-catch6
onehot-encoding6
Programming languages (10)
Github contributions (5)
FluxML/Flux.jl
Feb 2021 - Feb 2021
Relax! Flux is the ML library that doesn't make you tensor
Role in this project:
ML Engineer
ML EngineerContributions:18 reviews, 39 commits, 1 PR in 6 days
Contributions summary:Som primarily contributed to the implementation and testing of an orthogonal initialization function within the Flux.jl library. They modified the `src/utils.jl` file, adding the `orthogonal_init` function, modifying its implementation and examples, and also added tests for it in `test/utils.jl`. Through the commits, the user addressed bugs, improved code clarity, and expanded documentation. The user's work directly involved the creation of a key feature, including several iterations of improvements, showing active development in machine learning initialization techniques within the Flux framework.
ml-librarythe-human-braindata-sciencedeep-learningneural-networks
JuliaMath/Interpolations.jl
Jul 2021 - Jul 2021
Fast, continuous interpolation of discrete datasets in Julia
Role in this project:QA Engineer / Test Automation Engineer
QA Engineer / Test Automation EngineerContributions:8 reviews, 11 commits, 2 PRs in 28 days
Contributions summary:Som focused on enhancing the testing framework for the `interpolations.jl` library. Their commits include adding comprehensive unit tests for gradient calculations, particularly for extrapolation methods. The user improved test coverage, adding tests for various boundary conditions and data types, ultimately improving the reliability of the interpolation functionality. They also addressed inference issues, ensuring the type stability of the gradient calculations.
splinescontinuousdiscreteinterpolationjulia
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.