Skip to content
View BVishal-Geek's full-sized avatar
🌏
Master's Data Science at George Washington University
🌏
Master's Data Science at George Washington University
12 followers · 17 following

Achievements

Achievement: QuickdrawAchievement: Pull Shark

Achievements

Achievement: QuickdrawAchievement: Pull Shark

Block or report BVishal-Geek

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 user’s behavior. Learn more about reporting abuse.

Report abuse
BVishal-Geek/README.md

Hi, I'm Vishal Bakshi

I'm a Data Scientist with experience in AI/ML engineering specializing healthcare AI, insurance, and financial domains. I spend most of my time building intelligent systems that solve real-world problems. Previously, I built production AI at FCCI Insurance Group and conducted research at GWU's HIVE Lab.

What I'm working on

Currently, I'm focused on:

  • Building healthcare and financial models for real-world applications
  • Developing LLM pipelines for production systems
  • Creating conversational agents using GCP Dialogflow CX

I write about what I learn on LinkedIn. Recent posts:

Date Post
Day 1 Your model is deployed. But how does the user actually use it?
Day 2 "But it works on my machine!"
Day 3 Dockerizing Training vs Inference Pipelines and how inference actually goes live on AWS
Day 4 We ran containers on EC2 Day 4: We stop babysitting servers - Welcome to ECS
Day 5 What actually happens when GitHub Actions deploys an ML system to AWS ECS
Day 6 Observability vs Drift Detection - What's the difference and why it matters
Day 7 Understanding Drift in ML Systems (Drift is not about predictions. It's about what clicked for me)
Day 8 Your model drifted. Now what actually happens in production?
Day 9 Retraining is done. Now what? (This is where real ML engineering starts)
Day 10 Your model is deployed. But how does the user actually use it?

Open source

I've contributed to tools across ML infrastructure, computer vision, and data science workflows:

  • InsightFace: Enhanced face recognition pipeline error handling and robustness for production deployments

Research

Enhancing Prediabetes Diagnosis from Continuous Glucose Monitoring Data via Iterative Label Cleaning and Deep Learning of Bridge2AI AI-READI Data (Under Review)

Developing advanced deep learning techniques for early prediabetes detection using continuous glucose monitoring data, focusing on automated label cleaning and robust classification methods.

📫 Let's connect: LinkedIn

Pinned Loading

  1. NLP_Analysis_Airbnb_Reviews NLP_Analysis_Airbnb_Reviews Public

    Jupyter Notebook 1

  2. Deepfake_Video_Classification Deepfake_Video_Classification Public

    This project is a deepfake detection system built using TensorFlow, MLflow, and Streamlit. It allows users to upload videos through a web interface and get predictions on whether the video is real …

    Jupyter Notebook

  3. leetcode leetcode Public

    This repository contains my every day leet code submission.

    Python 1

  4. Legal_Document_Analysis_using_LLM Legal_Document_Analysis_using_LLM Public

    Forked from SidNadadhur/Legal_Document_Analysis_using_LLM

    AI Meets Law: Transforming Legal Research with RAG, Fine-Tuning, and Few-Shot Learning

    Jupyter Notebook

  5. Ollama_chatbot_pipeline Ollama_chatbot_pipeline Public

    This project builds an LLM-driven pipeline to evaluate biomedical research papers for machine-learning readiness. Given a paper PDF, it extracts text and applies strict, rule-based criteria to retu…

    Python 1

  6. deepinsight/insightface deepinsight/insightface Public

    State-of-the-art 2D and 3D Face Analysis Project

    Python 28k 5.9k