Skip to content
View Sahil-Chavan's full-sized avatar
๐Ÿ’ป
Eat.Sleep.Code.Repeat
๐Ÿ’ป
Eat.Sleep.Code.Repeat

Block or report Sahil-Chavan

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. Markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this userโ€™s behavior. Learn more about reporting abuse.

Report abuse
Sahil-Chavan/README.md

Header

๐Ÿ“– About Me

GIF

  • ๐Ÿ‘จ๐Ÿผโ€๐Ÿ’ป Data Scientist and AI/ML Researcher with a focus on building and deploying scalable, real-world solutions.
  • ๐Ÿ‘จโ€๐ŸŽ“ M.S. in Data Science @ Columbia University, NYC (Expected Dec 2025).
  • ๐Ÿ”ญ Current Research: Applying AI to single-cell genomics and immunology at Columbia's Mailman School.
  • ๐Ÿ‘จ๐Ÿผโ€๐Ÿซ Teaching Assistant @ Columbia, including Modern Data Structures and Advanced Machine Learning.
  • ๐ŸŒฑ Always learning and exploring the evolving field of Data Science.
  • ๐Ÿ‘ฏ Open to collaborations on research projects or industry-grade ML solutions.
  • โค๏ธ Interests: Music ๐ŸŽต, Coffee โ˜•, Movies ๐ŸŽฅ, and Documentaries ๐Ÿ“š.
  • ๐Ÿ“ My Resume

๐Ÿ’ก Key Experience & Focus

โžข โ˜๏ธ Cloud Research | Data Science & Gen AI Researcher

  • Architecting a state-of-the-art Generative AI feature to create synthetic AI personas for survey responses using Retrieval-Augmented Generation (RAG) and finetuned models.

โžข ๐Ÿš€ Amazon | Data Science Intern

  • Architected an autonomous agentic LLM platform for customer experience (CX) defect discovery, achieving a 100x velocity improvement over manual methods.
  • Built production-scale RAG pipelines using Amazon Bedrock Knowledge Bases for deep contextual grounding of CX violations.

โžข ๐Ÿข TIAA | Data Scientist

  • Engineered hybrid recommendation systems (XGBoost, BERT4Rec) that boosted advisor-led upsell/cross-sell rates by 18% and achieved 82% precision.
  • Built and managed scalable ETL pipelines over 10+ TB of data using AWS (Glue, S3), Snowflake, and PySpark.

๐ŸŽ“ Academic & Teaching Highlights

โžข Mentorship & Teaching: As a Teaching Assistant at Columbia, I have designed course content and mentored over 40 graduate students on projects spanning NLP, Computer Vision, and time series forecasting.

โžข Research & Publications: Authored two research papers published in Springer on a novel ensemble algorithm for object detection ("Non-linear Maximal Weighted Box Integration") developed during my undergraduate thesis.

โžข Early Internships & Impact: Gained foundational experience through internships where I developed award-winning NLP and recommendation systems that earned a government copyright and media recognition.

๐Ÿง‘๐Ÿฝโ€๐Ÿฆฒ Me and My Personality

โžข I believe in Smartness, Consistency, and Hard Work as keys to success.
โžข Passionate about sharing knowledge through mentoring and public speaking engagements.
โžข Actively involved in community service, contributing through tree plantations, educational drives, and cultural activities.

๐Ÿ’ป Technology & ๐Ÿ› ๏ธ Tools

Gen AI & NLP

ML/DL Frameworks

Data & Cloud Platforms

Web Dev & Deployment

๐Ÿ“ซ Reach out to me

Check out my Repositories ๐Ÿ—ƒ๏ธ

Pinned Loading

  1. Face_Recognition_Attendance_System Face_Recognition_Attendance_System Public

    Python

  2. Vehicle_Object_Detection Vehicle_Object_Detection Public

    Jupyter Notebook 1

  3. Automatic_License_Plate_Recognition Automatic_License_Plate_Recognition Public

    Jupyter Notebook

  4. Taxi_Prediction-CaseStudy Taxi_Prediction-CaseStudy Public

    Jupyter Notebook

  5. PersonalizedCancerPrediction_CaseStudy PersonalizedCancerPrediction_CaseStudy Public

    Here I have tried to solve an problem by Quora, posted on Kaggle. Problem Statement :--A lot has been said during the past several years about how precision medicine and, more concretely, how genetโ€ฆ

    Jupyter Notebook

  6. Stackoverflow_Tag_Prediction Stackoverflow_Tag_Prediction Public

    Problem Statemtent - Suggest the tags based on the content that was there in the question posted on Stackoverflow. Source-Kaggle(https://www.kaggle.com/c/facebook-recruiting-iii-keyword-extraction/โ€ฆ

    Jupyter Notebook 1