Skip to content
View sanjay-dilip's full-sized avatar
5 followers · 13 following

Achievements

Achievement: Pull Sharkx2Achievement: QuickdrawAchievement: Pair ExtraordinaireAchievement: YOLO

Achievements

Achievement: Pull Sharkx2Achievement: QuickdrawAchievement: Pair ExtraordinaireAchievement: YOLO

Highlights

  • Pro

Block or report sanjay-dilip

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 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
sanjay-dilip/README.md

Hi, I'm Sanjay.

I build end-to-end analytics systems - clean data in, dashboards and decisions out. Most of my work sits at the intersection of BI (Power BI), analytics engineering (SQL + Python), and applied ML when it adds real value.

If you only click two things: start with Levi's RAG (deployed AI due diligence copilot) and NBA Win Probability Engine (deployed Streamlit).

  • Levi's RAG - AI due diligence copilot, retrieval-grounded answers over SEC filings, deployed FastAPI + Next.js (Repo)
  • NBA Win Probability Engine - pre-game and live win-probability model, deployed Streamlit dashboard (Repo)
  • Sim2Real Engagement - sim vs real churn signals, Streamlit comparison dashboard (Repo)
  • MovieLens Recommender - ALS + hybrid ranking, deployed Streamlit app (Repo)

What I work on

  • DA/BI: Power BI (DAX), Tableau, KPI reporting, dashboard storytelling
  • Analytics: SQL, Python (pandas), experimentation, forecasting
  • Data Engineering: ETL/ELT, curated layers (Parquet), data modeling, quality checks
  • ML/AI: NLP, recommenders/ranking, RAG pipelines, LLM agents, model evaluation, deployment (FastAPI, Streamlit)

Highlights

  • I like projects where metrics tie to real decisions (late delivery risk, churn risk, ranking quality).
  • I care about reproducibility: clear READMEs, runnable steps, and basic checks.

Current Focus

  • Building retrieval-grounded and agentic AI systems with evaluation checks, not just demos (Levi's RAG, InsightPilot)
  • Shipping full ML apps end-to-end: model, evaluation, and a live deployed dashboard (NBA Win Probability Engine)
  • Strengthening data engineering habits: curated layers, data checks, and clean project structure
  • Keeping project READMEs and CI (GitHub Actions) current as each project ships

Tools I use

SQL, Power BI, Excel, Python (pandas, scikit-learn), Streamlit, FastAPI, LLM APIs (OpenAI, Gemini), RAG/vector search, Snowflake, Azure (ADF/Blob), GitHub Actions

Ongoing Projects

  • WC2026 Format Evaluation - assessing whether FIFA's 48-team World Cup expansion changes global representation, competitive balance, and scheduling fairness versus prior formats. Work in progress (Repo)

Contact

Outside of work, I like digging into sports, film, and world-event datasets, and increasingly building small AI agents to help make sense of them.

Pinned Loading

  1. levis-rag levis-rag Public

    Levi's RAG is an AI due diligence copilot for equity research on Levi Strauss & Co. It answers natural-language financial questions using only official SEC filings, grounding each answer in the exa…

    Python 1

  2. nba-win-probability-engine nba-win-probability-engine Public

    Machine learning dashboard for NBA win probability, covering pre-game predictions, live replay, model evaluation, and automated Finals refreshes with Google Cloud.

    Python

  3. insightpilot insightpilot Public

    An AI data analyst agent that turns raw CSVs into structured business insight reports — pandas handles all numeric analysis, Gemini handles only language, with guardrails and a 7-check evaluator t…

    Python 1

  4. movielens-recommender-system movielens-recommender-system Public

    End-to-end MovieLens recommendation system using ALS, hybrid ranking, evaluation pipelines, and a Streamlit interface.

    Jupyter Notebook

  5. supply-chain-disruption-analytics supply-chain-disruption-analytics Public

    Azure-based supply chain analytics pipeline that validates delay drivers statistically and predicts order-time delay risk to prioritize intervention capacity.

    Python

  6. sim2real-engagement sim2real-engagement Public

    Sim2real comparison of user engagement and churn modeling using simulated anime viewing data and real Steam gameplay data, highlighting how data constraints shape modeling decisions.

    Jupyter Notebook