Skip to main content
AboutProblemsGuidelinesTerms & conditions

Guidelines

These guidelines ensure a fair, productive, and enjoyable experience for all participants of the AjraSakha Hackathon.

Please read them carefully before participating.

Focus Areas & Requirements​

Technical Requirements​

Project Readiness​

  • The hackathon is driven by existing solutions of annam.ai and problem statements, and all solutions must be clean, deployable, and production-ready.
  • Solutions must be integrable into the Ajrasakha ecosystem and will be verified during evaluation.

Key Expectations:

  • Architecturally compatible with Ajrasakha’s tech stack
  • Simple, readable, and well-documented code
  • Deployable without major rework
  • You must understand and explain all submitted code

Bonus:
Teams that extend or improve the existing Ajrasakha system (instead of building parallel systems) will receive additional weight in judging.

Technology Stack​

Core (Required):

  • MongoDB, Express.js, React, Node.js

Allowed:

  • Python for AI/ML
  • OpenAI, HuggingFace, or custom models
  • OCR, translation, speech APIs
  • Vector databases
  • Maps APIs

Cost-Effectiveness (CRITICAL)​

All solutions must be viable for large-scale deployment in India.

Do:

  • Minimize GPU and API costs
  • Prefer open-source and cached data
  • Use expert-validated datasets
  • Design cheaper-than-human workflows

Don’t:

  • Use expensive APIs without justification
  • Depend heavily on paid LLMs
  • Require continuous GPU compute

Cost justification is evaluated in both phases.

Farmer-Centric Design​

Accessibility & UX:

  • Mobile-first
  • Works on low bandwidth
  • Supports at least one Indian language
  • Simple UI for low digital literacy
  • Voice support encouraged
  • Fast load times

Content:

  • Actionable farming advice
  • Expert-validated and localized
  • Region-specific recommendations

Two-Phase Monthly Structure​

Each hackathon runs every calendar month in two phases:

Phase 1: Build (2 Weeks)​

Objective: Build a complete working product

Deliverables:

  1. Source code (GitHub, MIT License)
  2. Working deployed application
  3. Demo video (2–5 mins)
  4. README & integration documentation

Notes:

  • Partial prototypes discouraged
  • All teams advance to Phase 2
  • Document planned Phase 2 improvements if incomplete

Phase 2: Improve & Refine (2 Weeks)​

Objective: Strengthen and polish Phase 1 product

Focus on:

  • Bug fixes and stability
  • UX and performance
  • Cost optimization
  • Stronger Ajrasakha integration
  • Presentation polish

Final evaluation considers both phases together.

Submission Requirements​

For both phases, submit:

1. Source Code​

  • Public GitHub repo (MIT License)
  • Clean structure
  • .env.example included

2. Working Application​

  • Live demo URL
  • Mobile compatible
  • Test credentials if required

3. Demo Video​

  • 2–5 minutes
  • Problem → solution → integration → impact
  • YouTube (unlisted) or Drive

4. Documentation​

  • README (setup, run, integration)
  • API docs (if any)
  • Cost breakdown

5. Optional​

  • Screenshots
  • Architecture diagrams
  • Sample data

Rules & Guidelines​

Do’s​

  • Build for Ajrasakha integration
  • Optimize for low bandwidth
  • Cache expensive operations
  • Test on mobile
  • Document clearly
  • Support regional language
  • Engage mentors

Don’ts​

  • Don’t ignore cost constraints
  • Don’t overcomplicate tech
  • Don’t submit non-working demos
  • Don’t skip documentation
  • Don’t assume high digital literacy
  • Don’t participate multiple times (rule may change in the future).

Intellectual Property​

You retain ownership of your work.

Repository:

  • Must be public
  • Must use MIT License

By participating, you allow annam.ai to:

  • Showcase your solution
  • Use screenshots, demos, and descriptions
  • Use team names for promotion and education

Conflict Resolution​

  • Mentor conflicts → decide best fit or ask organizers
  • Demo failure → show backup video
  • Late submission → rejected
  • Plagiarism → disqualification
  • Judging disputes → final decision
  • Integration failure → project considered invalid
  • Cost abuse → score penalty

Resources & Support​

Data Sources:

  • Agmarknet
  • eNAM
  • State agriculture portals

Infra:

  • Vercel / Netlify
  • MongoDB Atlas

AI/ML:

  • HuggingFace
  • Open-source models
  • Limited OpenAI usage

Point of Contact​

Lead Organizers​

Rajan Gupta
Role: Overall Coordination & Decision Making
Email: rajan.gupta@annam.ai

Aditya BMV
Role: Overall Coordination & Decision Making
Email: aditya.bmv@annam.ai

Technical Support Team​

Nandan Prabudesai
Role: Technical Support & Team Coordination
Email: nandanprabhudesai@annam.ai

Mamatha K
Role: Technical Support & Team Coordination
Email: mamathakuppireddigari@annam.ai

Bibin T
Role: Technical Support & Team Coordination
Email: bibin.t@annam.ai

Abiram K
Role: Technical Support & Team Coordination
Email: abiramk@annam.ai

Deepthi
Role: Technical Support & Team Coordination
Email: deepthi@annam.ai

Vatsal Khanna
Role: Technical Support & Team Coordination
Email: vatsalkhanna5@gmail.com

Ckesharwani
Role: Technical Support & Team Coordination
Email: ckesharwani4@gmail.com

Rishi
Role: Technical Support & Team Coordination
Email: rishia2220@gmail.com

Davinder
Role: Technical Support & Team Coordination
Email: davinder1436@gmail.com

Quick Contact Guide​

For General Queries:
Email any of the organizers