Abhishek Venkatachalam's Linkedin Analytics

Available Immediately to Join. Total Experience : 10 years Description : Data Scientist (LLM NLP NLU LRM - Semantic Searching) + Test Engineer SDLC/STLC exp: Agile Scrum, V model, Waterfall model. One of my Python Packages: https://pypi.org/project/easySemanticSearch/ (Easy Semantic Search based on SBERT and Sentence-Transformers for Quick, easy, multi threaded and efficient.) • Language: C# (DotNet), Python 3, VBScript • Data Science Environment: Google Cloud Platform, Jupyter Notebook, Google Colaboratory (Colab), VS Code • Machine Learning Libraries: Keras-RL, TensorFlow, PyTorch, Sentence-Transformers • Python Libraries: Numpy, Pandas, Scikit learn/ Sklearn, Seaborn, Matplotlib, ggplot2, SQL-DF, Google, Google-cloud, etc • Dev ops (CI/CD): Jira, Rally by CA, Azure ADO, GIT • Automation: MS Visual Studio (Selenium) , Provar, Pytest • Databases: MS SQL, Oracle, QMF • CRM: SalesForce Domain : Education, E-commerce, Telecom, Automobiles, Medicare financing, Banking Professional Summary: • Software Engineer at AXA Health UK, specializing in internal health insurance applications. • Developing an LLM Business, Technical and Process assistant based on RAG and Semantic Search (NLP) to assist internal engineers boost their productivity and efficiency. • Leading a data science project researching new methodologies to reduce context loss and improve memory retention in AI LLM using sophisticated embedding/ vectorization and recall. • Developing conversational chatbots with open-source SLM and RAG on power restricted CPU to improve performance for on-premise and on-device solutions that can be run without dedicated GPU. • Developing AI chat solutions with LLM models (GPT-3.5 and GPT-4) using Azure OpenAI, AI Services and other tools. • Developed a zero setup, easy to use semantic search python library based on SBERT, NLP, NLU and LRM and vector embeddings. (Package url: easySemanticSearch · PyPI) • Mentoring over 25 postgraduate students in Data Science and Machine Learning at Uniworx. • Advanced data analytics expertise in emissions from ports and vessels. • Led a 5-member team for a 3-month Machine Learning (ML) and forecasting project in collaboration with HSBC during my tenure at the University of Leicester. • A total of 9 years of experience across diverse software disciplines including Data Science, Data Analysis, Visualization and QA. • Gathered over 250,000 total downloads on a python package I built to easily obtain the entire Google Spreadsheet data with just the URL. (Package link: https://pypi.org/project/Googlexcel-noPassword/)