Lucky Verma
Machine Learning Scientist at Eccalon, LLC
Nashville-Davidson, Tennessee, United States
Summary
🤩
Rockstar🎓
Top SchoolLucky Verma is a Machine Learning Scientist with six years of experience building production-scale AI systems and a 4+ year focus on agentic AI platforms. He sole-architected AEGIS, an end-to-end system that cut foreign-ownership risk review from 40+ weeks to under 24 hours through knowledge-graph entity resolution across 10M+ corporate records and 95% ownership-chain accuracy. Lucky designs full-stack ML infrastructure—RAG pipelines, vector stores, FastAPI services, Docker/Terraform CI/CD—and has shipped automated code-security analysis and multi-path reasoning that materially improved retrieval relevance and review throughput. His background spans time-series biomedical research (PPG sleep and BP modeling), geospatial urban-mobility platforms, and recommendation/agent systems, showing versatility across domains and data types. Based in Nashville, he pairs hands-on engineering with product impact, translating complex compliance and industrial problems into auditable, deployable AI solutions.
6 years of coding experience
4 years of employment as a software developer
UMBCBachelor of Technology, Electrical and Electronics Engineering, Bachelor of Technology, Electrical and Electronics Engineering at SRM IST Chennai
Github Skills (91)
pre-trained-model8
stock-price8
generative-model8
keras-neural-networks8
extraction8
visualization8
foundation-models7
prediction7
mesh7
classify7
nlp7
multimodal-deep-learning7
language-understanding7
pressure7
vision-and-language7
Programming languages (6)
Github contributions (5)
lucky-verma/lucky-verma
Sep 2020 - Aug 2024
Contributions:42 pushes, 1 branch in 4 years
lucky-verma/Document-Classification-using-LayoutLM
Apr 2021 - Aug 2022
This PyTorch implementation of LayoutLM paper by Microsoft demonstrate the SequenceClassfication task using HuggingFaceTransformers to classify types of Documents.
Contributions:7 commits, 2 PRs, 5 pushes in 1 year 4 months
pytorchlayoutlmdeep-learningpytorch-implementationmicrosoft
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.