Adhikaar: India’s AI Welfare Copilot

AI-powered welfare copilot helping citizens discover government schemes, verify documents with OCR, and auto-fill applications using voice in their local language.

Comment

Inspiration

Millions of Indians miss out on government welfare schemes, not because they are ineligible, but because the process is confusing, fragmented, language-dependent, and often inaccessible. From rural citizens struggling with digital literacy to elderly people unable to navigate government portals, the welfare gap remains a major challenge.

We built Adhikaar to bridge this gap using AI. Our vision was simple: what if every citizen had a smart, multilingual welfare assistant that could understand documents, explain eligibility, and even help complete applications automatically using voice?

Adhikaar was inspired by the idea of making government welfare truly accessible, inclusive, and citizen-first.

What it does

Adhikaar is an AI-powered welfare copilot designed to help citizens easily access government schemes.

The platform helps users:

  • Discover eligible government schemes from a database of 4600+ welfare programs
  • Get personalized recommendations based on profile and eligibility
  • Interact using voice in local languages
  • Upload documents and extract information using OCR
  • Auto-fill government forms using AI-powered automation
  • Track welfare applications in one place
  • Access simplified explanations for complex government processes

Instead of forcing citizens to navigate multiple confusing portals, Adhikaar acts as a single intelligent assistant for welfare access.

How we built it

We built Adhikaar using a modern AI-first architecture focused on accessibility, multilingual interaction, and automation.

Frontend

  • Next.js 16
  • React 19
  • Tailwind CSS + shadcn/ui
  • Framer Motion + GSAP for smooth UI experience

AI & Intelligence Layer

  • Gemini for multimodal understanding and reasoning
  • OCR pipeline for document processing
  • Live voice interactions for conversational assistance
  • Eligibility matching system for scheme recommendations

Data & Automation

  • myScheme API integration for government welfare schemes
  • Playwright automation for smart form autofill
  • Document parsing and structured data extraction
  • Real-time user assistance workflows

We focused heavily on building a voice-first and low-friction experience to ensure accessibility for users with limited digital literacy.

Challenges we ran into

Building an AI welfare assistant for real-world public systems came with multiple challenges.

  • Understanding and structuring complex government scheme data
  • Handling multilingual and voice-first interactions reliably
  • Extracting accurate information from citizen documents
  • Managing dynamic government form structures for automation
  • Ensuring accessibility for users with limited technical knowledge

Creating an experience that feels simple for citizens while solving technically complex workflows was one of the biggest engineering challenges.

Accomplishments that we're proud of

We are proud of building a working welfare copilot that combines:

✅ Scheme discovery at scale
✅ Voice-first accessibility
✅ OCR-based document understanding
✅ AI-powered application autofill
✅ Personalized welfare assistance

Most importantly, we transformed a traditionally difficult government process into an experience that feels conversational, simple, and accessible.

What we learned

Through Adhikaar, we learned that impactful technology is not just about advanced AI, but about solving real human problems.

We gained experience in:

  • Multimodal AI systems
  • Voice-based user interaction
  • OCR and document intelligence
  • Automation for real-world workflows
  • Designing inclusive and accessible systems

This project reinforced our belief that AI can be used to create meaningful social impact at scale.

What's next for Adhikaar: India’s AI Welfare Copilot

Our next vision for Adhikaar is to evolve into a nationwide welfare operating system.

Future plans include:

  • Support for more Indian languages
  • Deeper integrations with government services
  • Real-time application status tracking
  • Offline-first experience for rural regions
  • WhatsApp and phone-call-based access
  • AI-powered grievance support

Our long-term mission is simple:

No citizen should miss government benefits because the system feels too complex to access.

Built With

  • docker
  • framer-motion
  • gemini-api
  • gsap
  • livekit
  • multimodal
  • myscheme-api
  • next.js-16
  • node.js
  • ocr
  • playwright-automation
  • python
  • react-19
  • rest-apis
  • sarvam-ai
  • shadcn/ui
  • tailwind-css
  • typescript
  • voice-ai
Share this project:

Updates