Skip to content
View khushijhanwar's full-sized avatar

Block or report khushijhanwar

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.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
khushijhanwar/README.md

Hi, I'm Khushi 👋

AI/ML Engineer and Data Scientist with expertise in Agentic AI, RAG,LLM and cloud data systems. Currently building AI browser-automation and content-automation systems as a Senior Consultant at Heartland Community Network. M.S. Data Science, Indiana University Bloomington (May 2026).

Contributed to four peer-reviewed publications across IEEE, Springer, and AIP, spanning VR feedback analysis, disease prediction, and speech recognition for Indian regional languages.

Contact · LinkedIn · Portfolio


Work Experience

Senior Consultant — AI Engineer · Heartland Community Network · Jun 2026 – Present : Building an AI browser-automation agent for read/verify/edit QA workflows on a live production CMS, plus automated content and internal-linking systems using semantic similarity.

Graduate Data Science Research Assistant · Indiana University Bloomington · Jan 2026 – May 2026 : Engineered reproducible ML data pipelines and a self-serve Streamlit reporting dashboard, cutting manual reporting effort 70%.

AI/ML Intern · Kellton Tech · Jun 2025 – Aug 2025 : Built an AI-powered resume-JD parsing system reaching 85% accuracy and benchmarked local vs. hosted LLMs on latency, accuracy, and cost.

Data Analyst · Atlas Copco, India · Jun 2023 – Jun 2024 : Analyzed telemetry data for early anomaly detection and deployed 20+ Nagios/Power BI dashboards, improving observability 60%.

ML Researcher · Vishwakarma Institute of Technology, Pune · Aug 2021 – May 2023 : Fine-tuned transformer models (Wav2Vec2, NVIDIA NeMo, BART) for multilingual speech recognition and ensemble learning models reaching 99% classification accuracy.


Projects

  • enterprise-ai-platform-v2 · Production-grade RAG + multi-agent platform on an open-source stack — DuckDB, FAISS, Ollama, LangGraph, FastMCP — with a two-stage retrieve-then-rerank pipeline and automatic fallback to extractive answers. Live demo
  • delivery-ops-analytics · SQL-based ELT pipeline transforming raw order-event data into canonical tables, with 7 automated data-quality checks and a self-serve Streamlit dashboard. Live demo
  • codebase-explainer · CLI tool that maps an unfamiliar codebase using parallel Claude Code subagents, one per module, merged into an architecture doc and onboarding guide. Tested against pallets/click. Live demo
  • Financial-Market-Insights-and-Analysis-Platform · Full-stack financial analytics platform for stock insights, volatility, news sentiment, and technical indicators. React, FastAPI, PostgreSQL.
  • MLB-research-project-IU · End-to-end ML pipeline demo, a synthetic reproduction of a real sports-analytics research project. pandas, scikit-learn, MLflow, Streamlit.

Education

M.S. Data Science · Indiana University Bloomington · Aug 2024 – May 2026

B.Tech. Information Technology · Vishwakarma Institute of Technology, Pune · Aug 2019 – May 2023


Tech Stack

Languages Python, SQL, Java, R, Bash, C, PHP

Agentic AI Google ADK, LangGraph, LangChain, MCP, FastMCP, multi-agent systems, tool calling, workflow orchestration, RAG, context engineering, conversational memory, prompt engineering

LLMs & NLP GPT-4, Claude, Gemini, Llama, Mistral, Ollama, TinyLlama, Hugging Face Transformers, FinBERT, embeddings, vector search, Pinecone, AlloyDB AI

Machine learning & deep learning TensorFlow, PyTorch, scikit-learn, XGBoost, SVD, TF-IDF, logistic regression, feature engineering, model evaluation, OCR

Data engineering & MLOps Spark, PySpark, Databricks, DuckDB, Pandas, MLflow, Docker, Kubernetes, CI/CD, Git

Cloud & databases GCP (AlloyDB, BigQuery, Dataflow), AWS (S3, RDS), Azure ML, PostgreSQL, MySQL, Oracle, MongoDB, Redis

Web development & APIs FastAPI, Flask, Node.js, React, Angular, REST APIs, JSON, Celery, Streamlit

Monitoring & BI Power BI, Tableau, Nagios, ServiceNow, Slack API, JIRA, SOP authoring, ISO 27001-aligned governance

AI-driven SEO & growth LLM-assisted keyword research & content classification, Google Search Console, Screaming Frog, Yoast SEO, technical/on-page audits, WooCommerce automation


Contact me: Email or LinkedIn

Pinned Loading

  1. enterprise-ai-platform-v2 enterprise-ai-platform-v2 Public

    Production-architecture RAG + AI agent platform on a free local stack — PySpark, DuckDB, FAISS, LangGraph, Ollama, FastMCP

    Python

  2. delivery-ops-analytics delivery-ops-analytics Public

    SQL-based ELT pipeline, automated data quality checks, and a self-serve dashboard for turning messy raw event data into trustworthy analytics.

    Python

  3. codebase-explainer codebase-explainer Public

    CLI tool that maps unfamiliar codebases using parallel Claude Code subagents — each explores one module in isolation, results merged into an architecture doc + onboarding guide. Tested against pall…

    Python

  4. Financial-Market-Insights-and-Analysis-Platform Financial-Market-Insights-and-Analysis-Platform Public

    A full-stack financial analytics platform for stock insights, volatility analysis, news sentiment, and technical indicators. Built with React, FastAPI, and PostgreSQL.

  5. jetstream2-mongo-project jetstream2-mongo-project Public

    A Jetstream2-based project that demonstrates the setup of a NoSQL MongoDB instance using Docker and the ingestion of airport data via Python. Includes cloud VM setup, data pipeline, PyMongo queries…

    Python

  6. MLB-research-project-IU MLB-research-project-IU Public

    End-to-end ML pipeline demo (pandas, scikit-learn, MLflow, Streamlit) — synthetic reproduction of a real sports-analytics research project.

    Jupyter Notebook