Websites & Web Apps
Business sites, internal dashboards, admin panels, booking or e-commerce flows — built clean and fast.
- React / Next.js or server-rendered
- FastAPI, Flask, or Node.js backends
- PostgreSQL / MySQL data layer
FULL-STACK DEVELOPER — PUNE, INDIA
I build systems that talk to each other.
Then I ship the interface.
Freelance developer building custom web apps, backend systems, and multi-agent AI pipelines — the kind that connect, reason, and act, not just chat.
0
PRODUCTION PROJECTS SHIPPED
0
PIPELINE, LIVE ON KAGGLE 2026
E2E
DATABASE TO DEPLOYED UI
§ 01 — SERVICES
Three ways I can help
What I actually deliver — picked to match what a business or founder needs, not a list of buzzwords.
AI Agents & Automation
Chatbots and agents that connect to your actual data — lead gen, support, outreach, internal workflows.
- RAG-powered assistants (ChromaDB)
- Multi-agent pipelines (Google ADK, CrewAI)
- Human-in-the-loop automation
Data & ML Tools
Turn raw data into decisions — prediction models, live dashboards, reporting people actually check.
- Streamlit or Power BI dashboards
- Predictive models & risk scoring
- Data cleaning & pipeline automation
↳ Need something outside these three? Node.js, Express, MongoDB, TypeScript and more are fair game — just ask.
§ 02 — SELECTED WORK
Recent builds
A few projects that show the range — from a multi-agent AI pipeline to a deployed ML API.
A 4-agent pipeline that researches companies, scores leads, and drafts personalized outreach in ~13 seconds. Standalone MCP server, long-term vector memory, prompt-injection defense, full audit logging. Built for the Kaggle AI Agents Competition 2026.
FIG. 02
Insurance Premium API
FastAPI
Scikit-learn
Docker
AWS EC2
A machine learning API predicting insurance premium categories from health, lifestyle, and demographic data — engineered features, Pydantic validation, an interactive Streamlit frontend, and a full Docker → EC2 deployment pipeline.
FIG. 03
Gmail Outreach Automation
CrewAI
Groq LLM
Supabase
A/B Testing
A multi-user email outreach platform sending up to 2,000 personalized emails/day — AI-generated templates, 48-hour reply/bounce detection, warm-up scheduling, and a 10-page analytics dashboard.
§ 03 — HOW IT WORKS
A straightforward process
No surprises. You'll always know what stage the project is in.
-
01
Discover
We talk through what you actually need — the outcome, not just the feature list.
-
02
Build
Staged builds with regular check-ins, so you see progress instead of waiting for a reveal.
-
03
Launch
Deployed, tested, and handed over — with documentation so it's never a black box.
-
04
Support
A free 2-week bug-fix window after launch, then ongoing support if you want it.
§ 04 — TECH STACK
What I build with
Core tools I use daily, plus a wider range I can pick up for the right project.
Frontend
- HTML / CSS / JavaScript
- React (working knowledge)
- Streamlit
Backend & APIs
- FastAPI
- Flask
- REST API design
- PostgreSQL / MySQL
AI & ML
- Google ADK / FastMCP
- ChromaDB (RAG)
- CrewAI, LangChain
- Scikit-learn, TensorFlow
Data & DevOps
- Pandas, NumPy
- Power BI, DAX
- Docker
- AWS (EC2, S3)
ALSO AVAILABLE ON REQUEST
Node.js
Express
Next.js
TypeScript
MongoDB
Google Cloud Platform
Django
§ 05 — MAINTENANCE & SUPPORT
Projects don't end at launch
Software needs upkeep. Here's how ongoing support works after your project ships.
BASIC RETAINER
Keep it running
- Monthly check-in & health review
- Security & dependency updates
- Small bug fixes
- Email support, 48-hr response
GROWTH RETAINER
Keep it improving
- Everything in Basic
- New feature additions each month
- Performance & uptime monitoring
- Priority turnaround, 24-hr response
↳ Every project includes a free 2-week bug-fix window after launch — no retainer required for that part.
§ 06 — GET IN TOUCH
Have something to build?
Tell me what you're trying to do — even a rough idea is enough to start the conversation.
Email me →