Logesh Umapathi
Machine Learning Consultant NLP, LLMs , Code Generation
Chennai, Tamil Nadu, India
Summary
👤
Senior🎓
Top SchoolLogesh Umapathi is a Machine Learning Consultant specializing in NLP, LLMs, and code generation with a decade of experience building ML solutions from idea to production. He has led teams and research efforts across enterprise and startup environments—from Tata Consultancy Services and Saama to BLACKBOX.AI—bridging applied research and product deployment. His open-source contributions include enhancing the BigCode evaluation harness to better assess autoregressive code-generation models, reflecting a practical focus on model evaluation and prompt engineering. Based in Chennai, he also mentors ML practitioners and publishes technical writing, blending hands-on engineering with knowledge sharing. An EEE graduate, he brings a rare mix of systems-level thinking and applied NLP expertise that drives production-ready LLM capabilities.
10 years of coding experience
8 years of employment as a software developer
B.E Electrical and Electronics, B.E Electrical and Electronics at R.M.K Engineering college
Github Skills (9)
transformers10
machine-learning10
evaluation-framework10
code-generation10
nlp10
python10
natural-language-processing10
autoregressive-models9
huggingface-transformers9
Programming languages (5)
Github contributions (5)
bigcode-project/bigcode-evaluation-harness
Nov 2022 - Jan 2023
A framework for the evaluation of autoregressive code generation language models.
Role in this project:
ML Engineer
ML EngineerContributions:6 reviews, 9 commits, 4 PRs in 1 month
Contributions summary:Logesh's commits focused on enhancing the evaluation framework for autoregressive code generation language models. They added a new task (CodeXGlue Text-to-Text) to the evaluation suite, including prompt formatting, generation settings, and post-processing functions. Furthermore, the user refactored and updated task-specific classes and registry configurations. These changes indicate a focus on expanding the evaluation capabilities of code generation models.
nlpcode-generationevaluationmachine-learningrefactoring
infinitylogesh/NLP_paper_notes
Nov 2021 - Nov 2022
Contributions:5 commits, 4 pushes, 1 branch in 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.