I'm Divy, a Data Science student who loves building software that actually gets deployed.
My interests sit at the intersection of three worlds:
- 🔬 Data Science: AI, ML and LLMs
- 📱 App Development: Flutter, Swift and Android
- ☁️ Cloud: AWS, DigitalOcean, Cloudflare and self-hosted infrastructure
I enjoy going beyond writing code, taking every project through:
Idea → Architecture → Development → Deployment → Monitoring
I've worked with cloud platforms such as AWS, DigitalOcean and Cloudflare, while also running my own Linux server infrastructure and self-hosting services.
My goal is to combine Data Science + Software Engineering + Cloud Infrastructure to build intelligent, scalable applications.
🤖 AI
LLM applications RAG systems Local / on-device AI AI-powered automation Machine learning pipelines
📱 Apps
Flutter Android iOS Full-stack web applications
🧩 Backend
REST APIs FastAPI Flask PostgreSQL MongoDB Firebase
☁️ Cloud
AWS Cloudflare DigitalOcean Docker Linux servers Self-hosted infrastructure
A private AI system designed to reduce everyday cognitive overhead.
Exploring LLMs, memory systems, RAG, automation, local inference and personal AI infrastructure.
Building tools that can analyze large datasets, identify quality problems and provide actionable insights without requiring users to manually inspect massive CSV files.
Experimenting with on-device AI, mobile applications and lightweight LLM inference.
From running models locally to integrating intelligence directly into apps.
I genuinely enjoy the "why pay for infrastructure when I can build it?" side of engineering.
Everything runs on a Linux server, with:
Docker PostgreSQL FastAPI Ollama / Local AI Cloudflare Tunnels Tailscale Self-hosted services
I like understanding what happens under the hood rather than treating deployment as a black box.
- 🤖 LLMs & Generative AI
- 📚 Retrieval Augmented Generation
- 📱 On-device AI
- ☁️ Cloud Infrastructure
- 🐳 Docker & Self-Hosting
- 🧠 AI Agents & Memory
- 📊 Data Engineering & Data Quality
- 🔬 ML Systems
Build it. Break it. Understand it. Deploy it.
I like projects where I get to work across the entire stack:
Data → Model → Backend → Application → Infrastructure → Deployment
Because the most interesting systems aren't built in just one layer.
