Skip to content
View eshan-chawla's full-sized avatar
πŸ“š
Learning
πŸ“š
Learning

Block or report eshan-chawla

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
ESHAN-CHAWLA/README.md

ESHAN CHAWLA

Machine Learning Engineer Β· AI Researcher Β· Backend Developer
Specializing in end-to-end development of applied AI solutions & cloud infrastructure πŸš€

Email Β Β  LinkedIn Β Β  GitHub Β Β  Website


🧭 Summary

Master’s student in Artificial Intelligence at San JosΓ© State University with hands-on experience across the full AI development lifecycle.

  • Expert in end-to-end development and deployment of AI solutions via backend services on cloud infrastructure.
  • Engineered RAG, OCR, and LLM workflows to automate core business operations and design robust data pipelines.
  • Experienced in Python, SQL, and Machine Learning with a foundation of academic excellence.

πŸ† Honors & Awards

πŸ₯ˆ 2nd Place @ Locus (YC F25) Agentic Payments Hackathon

San Francisco, CA

  • Built an AI Agent for Corporate Travel that automates booking logistics for sales teams and conferences.
  • The agent utilizes the Anthropic SDK and Kiwi MCP Server to pull real-time flight/hotel data, generates personalized itineraries, and negotiates adjustments with attendees via chat before auto-booking within budget.
  • Developed in just one day with a team of 4 (Uday Arora, Surya Pugazhenthi, Mukund P).
  • Prizes: $2,000 Stripe credits, Team Lunch at Stripe HQ, and Swag.

πŸ’Ό Work Experience

Machine Learning Engineer II @ PocketPills

πŸ“… July 2024 – July 2025

  • Optimized AI Service: Lowered Docker image build size by 42% and build time by 30% by replacing Chroma DB with AWS Bedrock Knowledge Bases.
  • Mentorship & Automation: Led the AI team in building RAG-based SMS response and OCR digitization workflows, saving operations 20 hours/week and $2,000/month.
  • System Integration: Reduced data-point consumption by 87% for Braze integration via a fault-tolerant API-based data ingestion pipeline.

Associate Data Analyst @ PocketPills

πŸ“… July 2023 – June 2024

  • Financial Impact: Decreased working capital by $300k by implementing Machine Learning time-series forecasting for just-in-time inventory management.
  • NLP Implementation: Developed an Inbound SMS classification system using BART embeddings clustering to automate message categorization.
  • Data Engineering: Engineered cost-efficient pipelines using AWS CDC to sync DB replicas to Databricks from sources like Mixpanel and PostgreSQL.

Data Analyst Intern @ PocketPills

πŸ“… Jan 2023 – June 2023

  • Automated reporting systems using Google Apps Script to sync Redash data to Google Sheets and Slack, eliminating manual entry.
  • Authored Tableau KPI reports to analyze team efficiency and call wait times for strategic decision-making.

πŸš€ Projects & Open Source

Contributor to the official implementation of "Forget-Me-Not" (CVPR 2024 Workshop).

  • Fixed critical configuration issues in the Textual Inversion (TI) training modules to ensure correct concept forgetting in diffusion models.
  • Tech Stack: Python, PyTorch, Diffusers.

Award-winning AI Agent designed to solve corporate travel logistics.

  • Functionality: Automates itinerary generation, personalizes bookings based on user preferences, and handles real-time negotiation and booking via Kiwi MCP.
  • Tech Stack: Anthropic SDK, Model Context Protocol (MCP), Python.

Production-grade RAG deployment on a user-facing portfolio site that allows users to chat with my professional experience.

  • Features: Implemented security configurations to prevent exploitation and optimized vector retrieval for low-latency responses.
  • Tech Stack: Next.js, Vercel, Pinecone, Vertex AI.

Machine Learning Security Solution to detect network anomalies and cyber threats.

  • Methodology: Implemented supervised learning algorithms on network traffic datasets (NSL-KDD) to classify normal vs. malicious activity.
  • Tech Stack: Python, Scikit-learn, Pandas.

NLP-based classification engine designed to segment and identify spam messages with high accuracy.

  • Methodology: Utilized Bag-of-Words models combined with Random Forest, Decision Tree, and Naive Bayes algorithms.
  • Tech Stack: Python, Jupyter Notebooks, NLTK.

πŸ›  Tech Stack

Languages

Python SQL JavaScript

ML / AI Frameworks

PyTorch TensorFlow Pandas NumPy Anthropic LangChain

DevOps & Cloud

Docker AWS GCP Vercel


πŸ“œ Certifications

  • NVIDIA: AI Infrastructure and Operations Fundamentals
  • Stanford Online: Machine Learning Specialization (Unsupervised Learning, RL)
  • DeepLearning.AI: Advanced Learning Algorithms & Supervised Machine Learning
  • MongoDB University: MongoDB Basics

πŸŽ“ Education

πŸŽ“ Master of Science (M.S.) in Artificial Intelligence San JosΓ© State University, San JosΓ©, CA, USA πŸ“… Aug 2025 – Present

πŸŽ“ Bachelor of Technology (B.Tech) in Computer Science Vellore Institute of Technology, Chennai, India πŸ“… July 2019 – June 2023


⭐️ From eshan-chawla

Pinned Loading

  1. SJSU-UNLEARNING-LAB/REDO SJSU-UNLEARNING-LAB/REDO Public

    Jupyter Notebook

  2. YC-HACK YC-HACK Public

    TypeScript 2 1

  3. ESHAN-CHAWLA ESHAN-CHAWLA Public

    1 1