About the role
About LightHouse:
LightHouse is an international, online hackathon for builders of agentic AI. Over several weeks, teams design and ship autonomous AI agents that plan, reason, use tools, and take action to solve real-world problems. Whether you're a student, a developer, or an early-stage founder, this is your chance to build, get noticed, and win.
Mandatory for Registration:
Registration is complete only when all team members do of the following:
- Fill the Profile Review Form: https://tally.so/r/1Wd0MM
Prizes:
- ₹10,000 in cash prizes
- ₹5,00,000 in kind (credits, tools, mentorship, perks, and opportunities from partners)
Format:
Fully online, with no travel required. Teams work on one problem statement from the released set and build an agentic AI solution.
Timeline:
Stage
Dates
Registration
13 Nov – 25 Dec
Round 1: Profile Review
15 Nov – 1 Dec
Problem Statements Released
1 Dec (announcements continue until 30 Dec)
Round 2: Build Phase (working on the problem statement)
1 Jan – 10 Jan
Final Submission
11 Jan
Grand Finale
Online (date and time to be announced)
Process:
- Register your team before the deadline.
- Your team's profiles go through a profile review.
- Shortlisted teams receive the problem statements.
- Pick one problem statement and build your agentic AI solution from 1 to 10 Jan.
- Submit your project on 11 Jan (working demo, GitHub repository, demo video, and a short write-up).
- Top teams present live at the online finale in front of judges.
Rules:
- All work must be created during the build phase; pre-existing code is allowed only if it is open-source and disclosed.
- Solutions must use an agentic AI approach (autonomous planning, tool use, or multi-step reasoning).
- Plagiarism or copied projects lead to disqualification.
- Any AI models, APIs, and frameworks are allowed, but you must credit them.
- Teams must submit before the deadline. Late submissions won't be considered.
- The organisers' and judges' decisions are final.
- Follow the code of conduct and behave respectfully with other participants.
Judging Criteria:
Innovation, agentic capability and autonomy, technical depth, real-world impact, usability, and presentation.
Join the community:
WhatsApp group for updates, doubts, and team-building: https://chat.whatsapp.com/JFwnM2kPMj6Dv7PY2K7OWA?mode=gi_t
Rounds to add:
- Profile Review (15 Nov – 1 Dec): Screening Round
- Problem Statement Release (1 Dec – 30 Dec): Announcement stage
- Build Phase (1 Jan – 10 Jan): Working on the problem statement
- Submission (11 Jan)
- Grand Finale (Online)
Eligibility
- Open to participants worldwide (students, professionals, and independent builders).
- Team size: 2 to 5 members.
- Inter-college and inter-specialization teams are allowed.
- Each participant can be part of only one team.
- Basic knowledge of programming and AI/LLM concepts is recommended.