Hunarmand worker

Inspiration

The inspiration behind Hunarmand Worker is personal. My father is a skilled worker, and I have seen how difficult it can be for skilled people to find the right customers and opportunities without formal qualifications or strong digital skills.

This made us realize that millions of skilled workers face the same problem. We wanted to use AI to connect their skills with real opportunities.

What it does

Hunarmand Worker is an AI-powered skilled-services marketplace. Customers can describe their needs using voice, text, or images, while AI agents understand the request and help find, match, and book the right skilled worker.

How we built it

We built it with Google Gemini + Google ADK, React, TypeScript, Node.js, Neon PostgreSQL, Drizzle ORM, Firebase, Cloudinary, Stripe, and OpenStreetMap + Leaflet. Customer and Worker agents use tools/function calling to interact with real workflows.

Challenges we ran into

Our biggest challenge was connecting AI reasoning with real-world workflows such as worker matching, booking, job states, payments, and authentication.

Accomplishments that we're proud of

We built an agentic marketplace with Customer and Worker AI agents, multimodal input, real database persistence, worker matching, booking, payments, maps, and notifications.

What we learned

We learned that agentic AI is more than a chatbot. Real agents need tools, data, permissions, structured workflows, and reliable actions.

What's next for Hunarmand Worker

We plan to improve multilingual voice AI, intelligent matching, agent memory, automation, trust features, and expansion to more cities and service categories.

“Har Hunar Ko Mauqa, Har Kaam Ke Liye Sahi Worker.”

Built With

• adk
• agentic
• agents
• ai
• calling
• cloudinary
• drizzle
• firebase
• full
• function
• gemini
• generative
• google
• leaflet.js
• multimodal
• neon
• node.js
• openstreetmap
• orm
• postgresql
• react
• stripe
• typescript
• voice