Manthan Thakker
Senior Software Engineer at Microsoft
Redmond, Washington, United States
Summary
👤
Senior🎓
Top SchoolManthan Thakker is a Software Engineer 2 at Microsoft with nine years of experience building production-grade cloud and AI services, currently contributing to Azure Cognitive Services on deep learning, NLP, speech and vision. He specializes in sequence-to-sequence models (Transformers, LSTMs), multi-task and transfer learning, knowledge distillation, and optimizations for real-time multilingual and multimodal scenarios. Earlier roles at Microsoft and startups span high-availability systems, cost-saving infrastructure optimizations, large-scale streaming data pipelines, and a lightweight versioning system that processed 600M records in minutes. As a former co-founder who launched a multi-state e-commerce site and a presenter of technical talks to Northeastern CS students, he blends product intuition with research-driven engineering. He holds an MS in Computer Science from Northeastern and a CS bachelor from the University of Mumbai, and is engaged with Microsoft Research initiatives toward integrative AI.
9 years of coding experience
5 years of employment as a software developer
Master’s Degree Computer Science, Master’s Degree Computer Science at Northeastern UniversityBachelor’s Degree Computer Science, Bachelor’s Degree Computer Science at University of Mumbai
English, Gujarati, Hindi, Marathi
Github Skills (42)
arima10
yahoo-finance10
recurrent-neural-networks10
perceptron10
reddit10
azure-storage9
feature-extraction9
speaker-identification9
fasttext9
classification9
bitcoin9
machine-learning9
mlp9
time-series9
backpropagation9
Programming languages (6)
Github contributions (5)
manthanthakker/speakerIdentificationNeuralNetworks
Nov 2016 - Jan 2020
⇨ The Speaker Recognition System consists of two phases, Feature Extraction and Recognition. ⇨ In the Extraction phase, the Speaker's voice is recorded and typical number of features are extracted to form a model. ⇨ During the Recognition phase, a speech sample is compared against a previously created voice print stored in the database. ⇨ The highlight of the system is that it can identify the Speaker's voice in a Multi-Speaker Environment too. Multi-layer Perceptron (MLP) Neural Network based on error back propagation training algorithm was used to train and test the system. ⇨ The system response time was 74 µs with an average efficiency of 95%.
Contributions:10 commits, 23 pushes, 1 branch in 3 years 2 months
backpropagationdatabasefeature-extractionmlpneural-network
manthanthakker/InformationRetrieval
Dec 2016 - Feb 2023
Contributions:9 commits, 8 pushes, 1 branch in 6 years 2 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.