# Divyanshu Tiwari

> Data Scientist and ML Systems Builder working across machine learning, data, AI applications, and software.

## About
 
22, working in Data & AI. Most days I'm building data pipelines and trying to make them actually reliable not just working once and breaking later. Lately I've been figuring out where LLMs and RAG genuinely help with data work, and where people are just forcing AI into things that don't need it. Outside of that: chess, coffee, and writing. Thanks for stopping by :)
 
- Location: India
- Email: divyanshutiwari281@gmail.com
- Website: https://datafolio.me/

## Projects

- [Customer Intelligence Engine](https://datafolio.me/projects): KMeans customer segmentation system with real-time cloud API on AWS EC2.
  - Category: Unsupervised ML & Cloud API
  - Stack: Python, Scikit-learn, KMeans, FastAPI, Docker, AWS EC2
  - GitHub: https://github.com/papocun/Customer-Segmentation-Project
  - Live Demo: http://16.171.71.103/

- [DemandCast](https://datafolio.me/projects): Production-grade bike demand forecasting pipeline with CatBoost and temporal validation.
  - Category: Time-Series & Production ML
  - Stack: Python, CatBoost, FastAPI, Docker, AWS
  - GitHub: https://github.com/papocun/DemandCast

- [Inventory Intelligence Engine](https://datafolio.me/projects): SQL-based ABC/Pareto inventory classification engine optimizing revenue-driving SKUs.
  - Category: Analytics & SQL Optimization
  - Stack: Python, Pandas, MySQL, Plotly, Streamlit, Railway
  - GitHub: https://github.com/papocun/Inventory-supply-chain-optimization

## Experience

- [Zeitster](https://datafolio.me/experience) (Aug 2026 – Present) — AI Data Engineer Intern
  - Designed end-to-end data infrastructure for ingesting Shopify and external merchant data covering GraphQL Bulk Operations, Webhooks, S3, Lambda, SQS, Redshift, Aurora PostgreSQL and DynamoDB.
  - Ingestion architecture for historical backfills and incremental data with retries, DLQs, idempotency and reconciliation.
  - Stack: AWS (S3, Lambda, SQS, Redshift), Aurora PostgreSQL, DynamoDB, Python, SQL, Shopify GraphQL, Data Modeling, Data Pipelines.

- [ChatSpark](https://datafolio.me/experience) (Mar 2026 – May 2026) — Data Analyst Intern
  - Automated Python Excel reporting pipeline end-to-end (cut generation time from 3+ hours to <5 mins).
  - SQL cohort analysis uncovering 15% decline in 90-day repeat purchases.
  - Stack: Python, SQL, Excel Automation, Cohort Analysis, Dashboards.

## Code

- [LeetCode](https://leetcode.com/u/21_dvynshx/): 130+ algorithm and data structure problems solved.
- [DailySQL](https://dailysql.in/u/divyanshutiwari281): 100+ SQL query optimization problems solved.
- [StrataScratch](https://platform.stratascratch.com/user/papocun): 80+ data science & query challenges solved.
- 1st Place Champion: Query Quest SQL Championship (500+ participants).

## Blog

- [Blog](https://datafolio.me/blog): Technical writing and research notes on ML systems, data engineering, and LLM applications (under construction).

## Contact

- Homepage: https://datafolio.me/
- Email: divyanshutiwari281@gmail.com
- GitHub: https://github.com/papocun
- LinkedIn: https://www.linkedin.com/in/divyansh21/
- Twitter / X: https://x.com/21dvy_t
- Resume: https://drive.google.com/file/d/14YWbiCTzkLcnwULHuuDESMvNvYIEjLbn/view?usp=sharing