Available for freelance & AI roles
I build AI agents that know when to ask a human.
I'm Gurleen Singh Bhatia — an AI engineer and MSc Data Science student specialising in agentic workflows, LangGraph orchestration, RAG architectures and Model Context Protocol integration. Published researcher on sub-10B autonomous agents, shipping FastAPI services with confidence-gated escalation that keeps humans in the loop.
- 0
- AI agent systems shipped
- >80%
- Task completion, sub-10B agents (published)
- MCP
- + LangGraph orchestration
- MSc
- Data Science — Chandigarh University
Featured work
Three systems, one obsession: agents that behave predictably.
AI Support Agent
Autonomous customer support that knows when to stop guessing.
An enterprise-grade multi-agent customer support and ticketing system for Autodesk-style workflows. A LangGraph StateGraph router performs zero-shot intent classification, sentiment and urgency triage, then hands off to specialised domain agents — escalating to humans over Redis Pub/Sub the moment confidence drops below 0.70.
- Python
- FastAPI
- LangGraph
- Streamlit
- Multi-Agent Systems
VideoMind
Transforming passive YouTube watching into active, high-retention learning.
A self-hosted, AI-powered study system that transforms YouTube videos and playlists into highly structured study notes. Integrates Firebase Google Auth, Firestore DB, Firebase Cloud Functions, and TanStack Router/Query with a Gemini 3.5 Flash prompt orchestration pipeline to generate executive summaries, handwritten-style notes, flashcards, timelines, and Q&A grounded in transcript text.
- React 19
- TypeScript
- TanStack Router
- TanStack Query
- Firebase Functions
Inbox Agent
An email assistant with a human still holding the pen.
An AI-powered professional email assistant that classifies incoming mail, drafts replies, and schedules meetings against the visitor's own Google Calendar — with a persistent human-in-the-loop review pipeline built on LangGraph interrupts.
- Python
- LangGraph
- Streamlit
- Groq API
- Google Calendar
Image Descriptor
One image in, publish-ready content out.
A multi-modal vision application built on Streamlit and the official Google GenAI SDK that turns any uploaded image into structured, professional content — blog posts, per-platform social copy, accessibility ALT text and SEO metadata, e-commerce listings, and interactive visual Q&A.
- Python
- Streamlit
- Google GenAI SDK
- Gemini
- Computer Vision
InvoiceWise OCR
Extracting structured ledger data from physical purchase receipts with OCR and AI.
A billing data extraction utility that processes scanned images of shop purchase bills. Utilizes the Tesseract OCR engine to read raw receipt text, structures it into formatted records, compiles it into downloadable Excel spreadsheets, and automatically pushes the extracted details to Zoho Books via API. Synced items and processing histories are backed up securely using Firestore database events.
- TypeScript
- Tesseract OCR
- Excel Generation
- Zoho API
- Firebase Auth
AI POS System
Point-of-Sale terminal automating item entry and ticket billing using AI.
A modern Point-of-Sale terminal designed to automate checkout billing for retail environments. Employs a TypeScript frontend for smooth terminal UI interactions, calling Firebase Cloud Functions as the serverless backend orchestration layer. Uses AI models to automatically categorize line items, predict billing totals, apply discounts, and store transactional records securely in the Firebase ecosystem.
- TypeScript
- React
- Firebase Functions
- Firestore DB
- Google Gemini API
What I can build for you
Freelance AI engineering, scoped around outcomes.
Pick the closest fit and the contact form arrives pre-filled with the context — no awkward “so what do you actually do” first message.
Custom AI Agent Development
Multi-agent systems that actually route, decide and act — not a single prompt in a loop. Built as a stateful graph with explicit intent classification, sub-agent handoff and confidence-gated fallbacks.
- Agent graph design
- Tool + API integration
- Eval harness & demo
RAG Chatbot Builds
Retrieval-augmented assistants grounded in your own docs, policies or product data. Chunking, embedding and retrieval tuned so answers cite reality instead of hallucinating it.
- Ingestion pipeline
- Vector store + retrieval tuning
- Chat UI or API
AI Support & Ticketing Automation
The system I've already built end to end: intent triage, sentiment and urgency scoring, autonomous resolution for the easy 70%, and clean context-rich escalation to humans for the rest.
- Intent & priority triage
- Resolution + escalation engines
- Ops dashboard
AI Automation Consulting
Short, blunt engagements to figure out where AI genuinely pays off in your workflow — and where it's an expensive detour. Architecture review, model selection, cost and latency planning.
- Workflow audit
- Architecture & model review
- Build roadmap
Live from GitHub
More things I'm building
Got a workflow that should be running itself?
Tell me what's slow, repetitive or drowning your team. I'll tell you honestly whether AI is the right tool for it.