class ShaanSatsangi:
def __init__(self):
self.name = "Shaan Satsangi"
self.role = "B.Tech CSE @ JECRC, Jaipur (Class of 2026)"
self.focus = ["Data Engineering", "AI Systems", "Analytics", "Full Stack"]
self.stack = ["Python", "SQL", "PySpark", "Databricks", "dbt", "Airflow", "FastAPI", "Next.js"]
self.currently = "Building production-grade data platforms, local-first AI tools, and develop intelligence platforms"
self.open_to = "Data Engineering / Analytics / AI Systems roles"
def say_hi(self):
print("Thanks for stopping by! Let's build something useful.")
me = ShaanSatsangi()
me.say_hi()- 📝 Medallion Lakehouse Architecture in Practice — Building Bronze
$\rightarrow$ Silver$\rightarrow$ Gold pipelines on Delta Lake. - ⚙️ Containerized Orchestration — Scheduling dbt data marts with Apache Airflow and Docker.
- 🧠 Offline RAG without Cloud APIs — Fastembed ONNX CPU vector search with FAISS.
Note
💡 Community Discussions: Want to collaborate, ask questions, or share your own builds? Join our active developer community on GitHub Discussions across my flagship repos!
Have a question about Data Engineering, local AI models, or my career journey? Ask away!
💬 How do you build local RAG pipelines in jarvis-py? (asked by @Shaan-alpha)
We use fastembed ONNX models for local CPU embeddings and FAISS for similarity vector search. No cloud APIs or torch dependencies required!
Read full repository →
| Project | Stack | What it does |
|---|---|---|
| CRM Sales Warehouse | Python · PostgreSQL · Docker · dbt · Airflow · Power BI | End-to-end CRM analytics platform with ETL/ELT pipeline, star-schema warehouse, dbt tests, and 5-page executive dashboard |
| YouTube Wrapped | Databricks · PySpark · Delta Lake · FastAPI · Next.js | Personal YouTube watch-history product using medallion lakehouse architecture, Neon Postgres, deployed API, and analytics dashboard |
| JARVIS-PY | Python · Ollama · Vosk · openWakeWord · FAISS · fastembed | Local-first AI voice assistant with wake-word barge-in, online/offline STT, semantic memory, PDF RAG, tool-agent routing, reminders, and interruptible TTS |
| Face Sort Studio | Python · Flask · OpenCV DNN · SQLite · SSE | Local face-recognition photo organizer using YuNet + SFace embeddings, real-time progress streaming, and privacy-first local processing |
| Sahaara | Next.js · TypeScript · Supabase · Tailwind · Twilio · MapLibre | Safety-focused full-stack app with emergency SOS workflows, live location sharing, trusted contacts, and Twilio alerts |
| Skill Issue | FastAPI · Next.js · GitHub API · AI · Analytics | Upcoming GitHub intelligence platform for developer identity analysis, OSS signals, recruiter readiness, and personality-driven diagnostics |
Data Engineering & Analytics
AI / ML / Local AI Systems
Full Stack & Tools
More tools I use
- 📚 B.Tech CSE @ JECRC Jaipur (2022 – Present) · CGPA: 7.39
- 🧠 Upgraded JARVIS-PY into a local-first AI assistant with RAG, semantic memory, wake-word barge-in, and tool-agent routing
- 🏗️ Building Skill Issue — a GitHub intelligence platform that analyzes developer identity, OSS activity, engineering maturity, and recruiter readiness
- 🌱 Studying for the Databricks Certified Data Engineer Associate exam
- 🤝 Open to Data Engineering, Data Analytics, and AI Systems roles
| 🏆 Certification / Achievement | 🏢 Issuer | 📅 Year |
|---|---|---|
| Certified Data Science, AI/ML Engineer & Data Engineer |  |
2026 |
| Fundamentals of Machine Learning |  |
2023 |
| Fundamental AI Concepts | |
2023 |
| 🥈 2nd Place – IEEE Debate Competition |  |
— |
Welcome to my Visitor Wall! Click below to sign my guestbook. An automated GitHub Action will instantly grab your GitHub avatar and greeting and add you to the wall below!
| Visitor | Message | Date |
|---|---|---|
 Shaan Satsangi Maintainer |
"Welcome to my GitHub! Drop a note and connect with me." | May 16, 2026 |
A huge thank you to everyone supporting my research in local AI assistants and open-source data tools! Sponsoring helps cover testing hardware and model hosting.
Explore my Sponsor Tier Architecture & Perks (SPONSORS.md) to see rewards like priority PR reviews, monthly 1-on-1 architecture advisory sessions, and premium logo placement!
"I build data platforms, AI systems, and full-stack products that turn messy inputs into useful outcomes."
