Tal Baumel
Principal Researcher at Microsoft
Israel
Summary
👤
Senior🎓
Top SchoolTal Baumel is a Principal Researcher in NLP at Microsoft with 12 years of experience bridging academic rigor and production research across Ben-Gurion University, Yahoo Research, and Microsoft. He holds a PhD in Computer Science and a background in computational linguistics, with a research focus that includes automatic text summarization and practical attention mechanisms for neural models. Tal has a history of improving neural toolkits—contributing attention implementations and memory fixes to the well-known DyNet library—demonstrating his ability to optimize both algorithms and low-level system behavior. Based in Israel, he combines deep theoretical knowledge with hands-on engineering to move NLP models from prototype to robust implementations.
12 years of coding experience
13 years of employment as a software developer
Master's degree, Computational Linguistics, Master's degree, Computational Linguistics at Ben-Gurion UniversityDoctor of Philosophy (PhD), Computer Science, Doctor of Philosophy (PhD), Computer Science at Ben-Gurion University of the Negev
Hebrew, English
Stackoverflow
31reputation
5kreached
0answers
4questions
Github Skills (15)
attention-mechanism10
machine-learning10
lstm10
deep-learning10
dynet10
python10
pagerank6
pandas6
mapreduce6
networkx6
text-files6
hadoop6
java6
cprogramming-language5
c-language5
Programming languages (7)
Github contributions (5)
clab/dynet
Jun 2016 - Nov 2016
DyNet: The Dynamic Neural Network Toolkit
Role in this project:
ML Engineer
ML EngineerContributions:8 commits, 3 PRs, 37 comments in 5 months
Contributions summary:Tal implemented and refined an attention mechanism example within the DyNet toolkit. Their work involved modifying and updating the `attention.py` file, experimenting with different attention function implementations, and adjusting the training process to improve convergence. The user also addressed memory allocation limits in the underlying C++ code of the library. The commits demonstrate a focus on improving the efficiency and functionality of neural network components.
dynetdynamic-neural-networkdeep-learningneural-networksmachine-learning
talbaumel/MIMIC
Nov 2016 - Oct 2019
Contributions:16 commits, 15 pushes, 1 branch in 2 years 11 months
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.