Agentic
Commerce
Hackathon
Build the future where AI agents can transact.
AI agents are moving from answering questions to taking action. What happens when they can also participate in commerce?
Hackathon complete
See the Top 3- Intent
- Discover
- Decide
- Authorize
- Transact
- Verify

// Results · Monday 5 October 2026
Top 3
10 teams built agentic commerce systems in six hours. Congratulations to the winners, and thank you to every team that took part.
Took part? Tell us how it went- 01Grand prize
Ээжийн Тек
Mership Agent
A buying agent on your side. Say what you need and your budget in Mongolian; it compares stores, reads product reviews, and pools people buying the same item into one cheaper group order that everyone approves.
₮5,000,000
- 02Second place
Баян Дата Технологи
DEALгүүр
Ask by voice in Mongolian and it searches online stores and Facebook listings, negotiates the price with the seller's agent, and asks you to confirm before paying into a sandbox escrow.
₮3,000,000
- 03Third place
Gwenchana
Sagslay
Tell it your weekly food budget and what is already at home. Agents plan the menu, ask store agents for live prices over A2A, and build the cheapest basket with the shortest shopping route.
₮2,000,000
Award ceremony · 6 October 2026
Prizes are presented at Data + AI Forum 2026, ZÜ Ballroom.
The teams show the solutions they built to forum guests during the networking session.
Top 3 prizes
₮10,000,000 in cash prizes for the three winning teams.
Audience Award
Forum guests pick their favourite team.
Sponsor gifts
Gifts from our sponsors for the teams.
The 10 teams, in pitch order
- 1Euro 92
- 2hackt1v4te
- 3Gwenchana
- 4Ээжийн Тек
- 5BTS
- 6Nomad Route
- 7Баян Дата Технологи
- 8Pi
- 9SB
- 10king of the north



// Build day program
Program
Monday 5 October 2026 · 09:00 – 20:20
01Opening & preparation
- 09:00 – 10:00Arrival & registration
- 10:00 – 10:15Opening ceremony
- 10:15 – 10:30Challenge briefing
- 10:30 – 11:00Oyu Intelligence sessionOyu use cases, MCP
- 11:00 – 11:30MongoDB sessionMongoDB use cases
- 11:30 – 12:00Team setup, final Q&A & group photo
02Development
- 12:00 – 18:00HackathonTeam development, 6 hours
- 12:00 – 14:00LunchMG Steak, any time in this window
- 16:00 – 16:30Coffee break
03Presentations & closing
- 18:00Submission deadlineCode freeze
- 18:00 – 18:15Break & judges setup
- 18:15 – 19:35Final demos & judging8 minutes per team
- 19:35 – 20:00Judges deliberation
- 20:00 – 20:15Top 3 announced
- 20:15 – 20:20Closing
// Technical framework
Requirements
An agent that acts, not a chatbot. Every submission has to clear all five. A component included only to tick the box will read as exactly that.
- 01
Agentic behaviour
At least one agent that interprets a goal and takes multiple steps toward it, using an external tool, data source, service or another agent. A prompt-to-LLM-to-answer interface is not enough on its own.
- 02
MongoDB
MongoDB must play a meaningful role: agent memory, workflow state, catalog, orders, retrieval and vector search, or audit history. Not included only to tick the box.
- 03
MCP
Expose or consume tools and data through the Model Context Protocol: catalog search, inventory, checkout, scheduling, logistics or your simulated payment service.
- 04
A2A
Agent2Agent between independent agents that discover capabilities, exchange tasks or coordinate a transaction. Keep it meaningful, not decorative.
- 05
OyuLLM
Use OyuLLM, the Mongolian language model, inside the agent. The OyuLLM team is on Discord in #qa.
// What you build with
Bring your own stack,
simulate your own payments.
- No sandbox or API keys from us
- Build and simulate your own store and payment setup. No real money, cards, bank details or real purchases.
- OyuLLM
- Required in every project. The OyuLLM team answers questions on Discord in #qa.
- MongoDB
- Open-source MongoDB for the data, memory, catalog and audit layer.
- Your team repo
- A private repo at aai-mn/team-XX on GitHub. Accept the invite in your email before you start.
// Final presentation
Eight minutes
in front of the judges.
| Problem & solution | 1 min |
|---|---|
| Live product demonstration | 3 min |
| Architecture & technology | 1 min |
| Judge Q&A | 3 min |
Demos start at 18:15. Slides support the demo rather than replace it. The judges want to see the agent and the transaction flow working.
// The fine print
Rules
Payments are simulated throughout: sandbox only, no real money and no real card credentials at any point.
| Rule | Detail |
|---|---|
| Team size | Up to 3 participants |
| Places | 10 teams, selected from the registrations |
| Who can enter | Developers, designers, students, researchers, founders and builders |
| Build freedom | Any language, LLM, open-source model, agent framework, frontend or cloud platform |
| AI coding tools | Allowed |
| Build it yourselves | Libraries and templates are fine. No old product passed off as new, no outside people building it |
| Pre-existing work | Allowed only when clearly disclosed |
| Source code | Pushed to your private aai-mn/team-XX repo by 18:00. No new features after that |
| Ownership | Rights to new work built here go to AAI NGO. Your earlier code stays yours |
| No secrets | No work code, customer data or secret keys. Demos are public |
| Payments | Build and simulate your own. No real money, cards, bank details or purchases |
| Demo | Finalists must show a working product or technically credible prototype |
A working demo is required. Judges may ask you to explain your code.
- Everything below in your aai-mn/team-XX repo by 18:00
- Project name and one-paragraph description
- Team member information
- Source-code repository
- README with setup and run instructions
- Architecture diagram
- Explanation of the agent workflow
- Explanation of MongoDB usage
- Explanation of MCP, A2A or other open interoperability used
- 2–3 minute demo video
- Working demo for finalist judging
- Disclosure of pre-existing code, datasets, templates or major components
Agents transact on someone's behalf, so these are part of the judging conversation.
- Clear user intent and authorization before consequential actions
- Configurable constraints: budget, quantity, category, seller, timing, approval thresholds
- Human confirmation for actions the team considers high-impact
- Transparent transaction status and receipts
- Auditability of important agent actions
- Failure handling, cancellation or refund flow where relevant
// How you are scored
Six criteria, weighted. 60% of the score sits on the problem you chose, the agent behind it and how well you used OyuLLM and MongoDB. Three judges score every team and pick the Top 3.
Problem & usefulness
20%Clarity of the problem, relevance of the use case and value created.
Agentic architecture
20%Quality of reasoning, planning, tool use, state management and autonomy.
OyuLLM & MongoDB usage
20%Meaningful use of OyuLLM inside the agent, and the depth and technical quality of MongoDB usage.
Open interoperability
10%Meaningful MCP, A2A or related open-interface implementation.
Technical execution
10%Working system, engineering quality, reliability and completeness.
Innovation & demo
20%Originality, product thinking and clarity of the live demonstration.
- Total100%
Judges prioritise useful, coherent systems over unnecessary complexity. A reliable single-agent application with strong tool use and a clear transaction workflow may score higher than a complicated multi-agent system that does not work.
// Challenge scope
What counts as agentic commerce?
10 directions to build in. Pick one, or bring a concept of your own that meets the technical and safety requirements.
- 01
E-commerce
An agent searches a catalog, compares products against user constraints, selects an item and completes a sandbox checkout.
- 02
Travel
An agent creates an itinerary, evaluates transport or accommodation options and coordinates a simulated booking.
- 03
Food & local commerce
An agent builds an order based on preferences, budget and availability, then places a simulated order.
- 04
Subscriptions
An agent manages renewals, upgrades or cancellations based on user-defined rules.
- 05
Events & ticketing
An agent discovers events, selects tickets under defined constraints and completes a sandbox purchase.
- 06
SME / B2B procurement
A purchasing agent requests offers, compares suppliers and coordinates an approved order.
- 07
Digital services
An agent selects and purchases software, compute, data, API access or other digital services.
- 08
Marketplaces
Buyer and seller agents discover each other, exchange structured information and coordinate a transaction.
- 09
Gaming / virtual goods
An agent manages permitted purchases or resource acquisition inside a simulated environment.
- 10
Your own concept
Any other compelling agentic commerce use case that meets the technical and safety requirements.
// How it runs
Timeline
From registration to the award ceremony. The build day was Monday 5 October 2026, and prizes are presented on 6 October 2026.
Every participant registered individually, including each member of a team. Registration is closed.
10 teams were selected and notified by email and phone.
Opening, challenge briefing, Oyu Intelligence and MongoDB sessions, then team setup, Q&A and a group photo.
Six hours of building on site.
Code freeze at 18:00. All 10 teams demoed from 18:15, and the Top 3 were announced at 20:00.
Prizes, the Audience Award and sponsor gifts are presented at Data + AI Forum 2026, ZÜ Ballroom, where the teams show their solutions to forum guests during the networking session.
// Who is behind this
Organizer
- Novelsoft
- Applied AI Mongolia
Community
- Applied AI Mongolia
- Agentic AI Foundation Ulaanbaatar
Technology partner
- MongoDB
Sponsor

Applied AI Mongolia is an independent community organizer and not an official channel of the Agentic AI Foundation. AAIF and the AAIF logo design are registered trademarks of the Linux Foundation.