Multi-Language Welfare Scheme Finder
Budget / Salary₹12,500–37,500
TypeFreelance project
LocationRemote
Posted3 hours ago
I need a highly interactive web application that lets Indian citizens instantly discover every Central and State welfare scheme they qualify for. The core of the product is an AI-driven search that surfaces relevant schemes based on plain-language queries, so “student from Nagpur, OBC, annual income under ₹3 L” should immediately return the right scholarships, subsidies, and other benefits.
Key interface decisions
• The landing page must feel lively yet uncluttered, with language selection offered as a single, elegant dropdown. Switching between English, Hindi, and Marathi should never require extra clicks or reloads.
• Real-time feedback as users type (auto-suggested ministries, categories, or keywords) will keep the experience fluid.
• Results need smart grouping—by Central vs State, then by department—so users see the full landscape at a glance.
Essential feature set
• Scheme search (this is the heart of the build).
– Semantic search powered by an LLM or fine-tuned embeddings.
– Filters for age, gender, income bracket, occupation, disability, and state.
– Deep links to official government PDFs and application portals.
Although an eligibility calculator and user accounts may come later, right now I want you to concentrate on delivering the most accurate, lightning-fast scheme discovery engine and the interactive front end that showcases it.
Tech stack suggestion (flexible)
React or Vue for the UI, Node/Express or FastAPI for the backend, PostgreSQL for structured scheme data, and OpenAI or Cohere for natural-language search. If you know better tools that achieve the same snappy response, propose them.
Acceptance criteria
1. Users can pick a language from the dropdown and see every static label translated.
2. A single search returns relevant schemes in under two seconds for a 10k-record dataset.
3. Each scheme card shows title, summary, benefit amount/type, eligibility snippet, and “Apply” link.
4. Mobile-first design passes Lighthouse scores of 90+ for performance and accessibility.
5. Clear README with setup commands and API documentation.
Hand-off deliverables
• Source code in a private Git repo
• Seed database or CSV with at least 50 sample schemes
• Deployment script (Docker or CI workflow) to spin up on an AWS, GCP, or Render free tier
Let’s craft something citizens can rely on.
Key interface decisions
• The landing page must feel lively yet uncluttered, with language selection offered as a single, elegant dropdown. Switching between English, Hindi, and Marathi should never require extra clicks or reloads.
• Real-time feedback as users type (auto-suggested ministries, categories, or keywords) will keep the experience fluid.
• Results need smart grouping—by Central vs State, then by department—so users see the full landscape at a glance.
Essential feature set
• Scheme search (this is the heart of the build).
– Semantic search powered by an LLM or fine-tuned embeddings.
– Filters for age, gender, income bracket, occupation, disability, and state.
– Deep links to official government PDFs and application portals.
Although an eligibility calculator and user accounts may come later, right now I want you to concentrate on delivering the most accurate, lightning-fast scheme discovery engine and the interactive front end that showcases it.
Tech stack suggestion (flexible)
React or Vue for the UI, Node/Express or FastAPI for the backend, PostgreSQL for structured scheme data, and OpenAI or Cohere for natural-language search. If you know better tools that achieve the same snappy response, propose them.
Acceptance criteria
1. Users can pick a language from the dropdown and see every static label translated.
2. A single search returns relevant schemes in under two seconds for a 10k-record dataset.
3. Each scheme card shows title, summary, benefit amount/type, eligibility snippet, and “Apply” link.
4. Mobile-first design passes Lighthouse scores of 90+ for performance and accessibility.
5. Clear README with setup commands and API documentation.
Hand-off deliverables
• Source code in a private Git repo
• Seed database or CSV with at least 50 sample schemes
• Deployment script (Docker or CI workflow) to spin up on an AWS, GCP, or Render free tier
Let’s craft something citizens can rely on.
Apply on Freelancer →
Project sourced from Freelancer.com. Applications happen directly on the original platform — we never collect your data.