Vin T
Principal at Samsung Next
New York, New York, United States
Summary
🤩
Rockstar🎓
Top SchoolVin T is a Principal machine learning engineer based in New York with 11 years of experience applying ML to product and research problems across video, IoT, and wearable domains. At Samsung Next he has shipped models for saliency and scene classification in immersive video, anomaly detection for devices, and gesture recognition on smartwatches while also evaluating and investing in ML startups. A hands-on engineer and educator, he contributes practical tutorials to open-source AI tooling (notably Deeplake) and has improved core libraries like imagehash with tests and clearer documentation. He prefers clarity—both in understanding complex systems and in explaining them—which shows up in his blend of product work, developer-facing examples, and startup cofounding experience in medical imaging.
11 years of coding experience
2 years of employment as a software developer
Carnegie Mellon University
Stackoverflow
351reputation
72kreached
17answers
4questions
Github Skills (27)
unit-testing10
pytorch10
learn-ruby-on-rails10
python10
data-science10
user-manual10
machine-learning10
basics10
datasets10
deeplearning-ai10
deep-learning10
tensorflow10
jupyter-notebook10
image-processing9
visualization9
Programming languages (6)
Github contributions (5)
activeloopai/deeplake
Dec 2020 - Apr 2021
Database for AI. Store Vectors, Images, Texts, Videos, etc. Use with LLMs/LangChain. Store, query, version, & visualize any AI data. Stream data in real-time to PyTorch/TensorFlow. https://activeloop.ai
Role in this project:
ML Engineer
ML EngineerContributions:7 reviews, 68 commits, 87 PRs in 4 months
Contributions summary:Vin appears to be focused on building tutorials related to the `deeplake` library, demonstrating its usage for storing, manipulating, and visualizing data, with specific examples involving images and labels. Their primary contribution is the creation of a quick start tutorial and other tutorial files, showing how to interact with the library to store and process data. The commits include example code using `hub` for dataset creation, manipulation and the use of `matplotlib` for visualization, suggesting a user who is building out the practical examples and tutorials of the tool.
pythonreal-time-datadata-streamjupyter-notebookvisualize
JohannesBuchner/imagehash
Dec 2016 - May 2017
A Python Perceptual Image Hashing Module
Role in this project:Back-end Developer & QA Engineer
Back-end Developer & QA EngineerContributions:8 commits, 2 PRs, 2 comments in 4 months
Contributions summary:Vin focused on improving the `imagehash` library, primarily by adding comments to explain the code, particularly in the average hashing algorithm implementation. They also added unit tests and refactored existing code to check hash size arguments, addressing potential errors. Furthermore, the user ensured that the library followed image fingerprinting principles by explaining why cryptographic hashing algorithms are not suitable in the README.
hashingpythonperceptualimage-processingimage-hashing
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.