Turn Your Knowledge into an LLM‑Ready API - Hackathon @ Generation AI [AI Tinkerers - Paris]

Turn Your Knowledge into an LLM‑Ready API - Hackathon @ Generation AI

Dec
09
Tuesday
December 9th, 2025 from 8:30AM (CEST)
Dec
10
Wednesday
December 10th, 2025 until 5:55PM (CEST)
Address Info
Available on RSVP acceptance

Event Ended

This event has already taken place.

Attendees 79+ registered
Talks include improving LLM context with GraphRAG and isolating AI backends using private network security, and more. View Demos »

⚡ Turn Your Knowledge into an LLM‑Ready API at Paris Hackathon with Scalingo and Neo4j

From Tuesday, December 9 (11:30 - 18:00) to Wednesday, December 10, 2025 (08:45 – 18:00), Paris becomes the epicenter of context-integrated GraphRAG.
Join top builders, researchers, and engineers to prototype the next generation of Model Context Protocol (MCP) integrations for graphRAG databases — and explore how structured memory can unlock reasoning at scale.

(Banner) A promotional banner for the 'GenerationAI Hackathon' featuring white text overlaid on a dark blue, futuristic background that includes technical blueprint diagrams and classical architectural domes, with two silhouetted figures viewing the display. Text: GenerationAI Hackathon MCP Integration for Graph DBs AI TINKERERS - PARIS Modern, technical, high-contrast. | Colors: #0d1b46, #ffffff, #2498b8, #f2e0a1 Note: This image is a structured graphic designed to promote a specific event (a hackathon) using clear, hierarchical text blocks set against a thematic, illustrative background.

Space is limited and we’re screening for hands-on talent.
If you work with LLMs, RAG, or graph-based reasoning, secure your spot early — once full, the waitlist opens.


🧭 Theme

Model Context Protocol Integrations for GraphRAG
We’re bridging the gap between language models and structured data systems. Expect a full day of experimentation around:

  • Context persistence across graph nodes
  • Dynamic retrieval and reasoning from structured sources
  • Hybrid memory architectures (Graph + RAG)
  • MCP interfaces for multi-agent collaboration and explainability
  • Visual query builders and interpretable context trails

Whether you’re a backend engineer, data scientist, or AI tinkerer, bring your boldest ideas — we’ll connect you with mentors and teammates to make them real.


🗓️ Schedule

Day 1: Tuesday, December 9, 2025

Time Activity Notes
8:30 AM Breakfast @ Partner Village
9:00 AM Welcome and Opening Remarks @ Amphi
9:15 AM General Keynotes  
10:35 AM Coffee Break / Networking / Expo @ Partners Village
11:00 AM - 11:25 AM AI Tinkerers Paris Hackathon Launch @ Dattier 3
11:30 PM - 11:55 PM Neo4j Tech Talk: Introduction to GraphRAG Workshop  
12:00 PM - 12:25 PM Team Formation Hackathon portal/ Teams
12:30 PM - 12:55 PM Tinkering  
1:00 PM Lunch Break @ Partner Village
2:00 PM - 3:55 PM Tinkering @ Dattier 3
4:00 PM Coffee Break / Networking / Expo @ Partners Village
4:30 PM - 5:30 PM Tinkering Final push before evening
5:30 PM End of Hackathon Day 1  
6:00 PM Optional Builder Station (separate registration required) @ Holiburton school

Day 2: Wednesday, December 10, 2025

Time Activity Notes
8:45 AM Breakfast @ Partner Village  
9:15 AM - 10:30 AM Tinkering @ Dattier 3
10:50 AM Coffee Break / Networking / Expo @ Partners Village
11:20 AM - 12:45 PM Tinkering @ Dattier 3
12:50 PM - 2:00 PM Lunch Break @ Partner Village
2:00 - 2:30 PM Last chance to polish demos @ Dattier 3
2:30 PM - 3:15 PM Final submission Hackathon Portal/ Compete/ Submit Entry tab
3:15 PM - 3:50 PM Judging Hackathon Portal/ Voting Tool
4:00 PM Coffee Break / Networking / Expo @ Partner Village
4:30 PM - 4:50 PM Finalist Presentations & Jury Deliberation @ Partner Village
4:50 PM Winners Announced & Closing Ceremony Hackathon Close

Day 3 (optional): Thursday, December 11, 2025

See the official program of the conferences


🛠️ What to Bring

  • Laptop + charger (extension cords = instant hero)
  • Your favorite dev stack or graph database setup
  • MCP SDKs, APIs, or open connectors
  • Curiosity and a done-is-better-than-perfect mindset

🏆 Judging & Prizes

We’ll reward projects that are working, elegant, and context-smart.
Extra points for:

  • Seamless MCP ↔ Graph integration
  • Context-aware reasoning or dynamic retrieval
  • Real-world usefulness and open-source potential
  • Clear UX for developers and data teams

🏋️ Judging Criteria

Criterion Focus & High Score Example Scoring (1 to 5)
Working Prototype Focus: Is the solution functional, deployable (Scalingo friendly), and does it demonstrate the core concept? High Score (5): Flawless execution; a live demo running the GraphRAG pipeline end-to-end. 1: Project does not run or is entirely conceptual.
2: Major bugs prevent core functionality demonstration.
4: Runs, but has minor bugs or dependency issues.
5: Flawless execution and stable demonstration.
Theme Alignment & MCP Integration Focus: How deeply does the project implement the “Model Context Protocol Integrations for GraphRAG”? High Score (5): Achieves seamless, bidirectional MCP ↔ Graph integration, demonstrating advanced features like context persistence across nodes or multi-agent collaboration using structured memory. 1: Barely relates to LLMs or graphs.
2: Superficial use of RAG or a simple graph structure.
4: Clear, meaningful incorporation of GraphRAG concepts.
5: Perfectly embodies and elevates MCP integration with Neo4j/Graph data.
Contextual Reasoning & Retrieval Focus: Does the project leverage the graph structure for smarter context retrieval and reasoning, moving beyond simple vector search? High Score (5): Implements dynamic, context-aware reasoning (e.g., pathfinding, relationship inference) that directly informs the LLM’s output, with clear explainability trails. 1: Retrieval is basic keyword/vector search only.
2: Graph is used only for static data storage.
4: Demonstrates dynamic retrieval based on a simple graph query.
5: Advanced, context-smart reasoning unlocks novel LLM capabilities.
Developer Experience (UX/DX) Focus: How usable is the resulting API/interface for developers or data teams? (As per judging notes: “Clear UX for developers and data teams”). High Score (5): Provides an elegant, well-documented API wrapper (Scalingo deployable) or a visual query builder that makes complex graph context easy to consume. 1: No clear interface or documentation.
2: Usable only by the original hacker.
4: Clear API endpoints or a simple interface exists.
5: Elegant, well-documented, and immediately useful for other developers/data teams.

🏆 Prizes

We recognize excellence in GraphRAG and MCP integration!

Neo4j Credits: The top 3 winning teams each receive 1,000 credits.

Scalingo Credits: The overall winning team and the winners of the Scalingo track each receive a €270 coupon code for the PaaS.

Speechmatics Credits: Each winner from each of the winning teams will each receive $200 in Speechmatics credits.

Google Cloud Credits: Each winner will receive $500 in Google Cloud credits, which can be applied to the billing account of their choice. These $500 credits are in addition to the standard $300 trial credits available to new Google Cloud accounts. These combined credits allow the use of Google Cloud resources, but a personal credit card must be linked to the account to activate and utilize the full credit amount. The total amount of credits is limited, as are the quotas (e.g., GPU access).


The Hackathon is part of Generation AI Paris 2025 — a 3-day explosion of innovation with 8,000+ attendees, 300+ sessions, and 100+ speakers.

You’re getting the full package: keynotes, track sessions, workshops… and all the networking breakfasts and lunches (because who codes on an empty stomach?).

But here’s the kicker: your acceptance isn’t just for the hackathon. It’s a golden pass to GenerationAI, apidays, AND GreenIO Paris. That’s 4 conferences — normally valued at 999€ — for the price of zero.

Yes… we said zero. You’re welcome.


Neo4j - Neo4j is a leading graph database platform that empowers AI developers to harness the power of connected data. Their technology enables efficient storage, analysis, and visualization of complex data relationships, integrating seamlessly with popular AI and machine learning tools. Neo4j’s features, including native graph storage and advanced algorithms, make it ideal for powering AI applications like knowledge graphs and recommendation systems, helping developers extract deeper insights from their data.

(Logo) The image displays the Neo4j logo, featuring the stylized text 'neo4j' in black, accompanied by a blue node-like graphic element to its left and a blue dot above the 'j'. Text: neo4j Colors: #000000, #005781, #FFFFFF Note: This image is a stylized graphical representation of a company or product name, designed for brand identification. It combines text with a unique visual element.

Scalingo is a European Platform as a Service (PaaS) designed to simplify cloud hosting and database management for developers. It enables rapid deployment, management, and scaling of web applications, supporting over 50 runtimes and integrating seamlessly with major databases like PostgreSQL, MySQL, and MongoDB. With features like auto-scaling, continuous deployment, real-time monitoring, and robust security, Scalingo empowers tech teams to focus on coding rather than infrastructure. New users can enjoy a 30-day free trial to experience its developer-friendly environment and top-notch support.

(Logo) A logo featuring the word "Scalingo" in dark gray text, accompanied by a graphic element of two slanted, rounded rectangular shapes in light blue and royal blue, all set against a white background. Text: Scalingo Colors: #262626, #33C7F5, #2243F5, #FFFFFF Note: The image clearly presents a stylized graphic element alongside a distinct brand name, designed for brand recognition, which are the defining characteristics of a logo.


🧠 Mentors

Samir AKARIOH

Samir AKARIOH

Developer Advocate @ Scalingo

Henry Collie

Henry Collie

Technical curriculum developer @ Neo4j

🙌 Volunteers

Hamze GHALEBI

Hamze GHALEBI

ex CTO @ Welcome Account


🔒 Logistics & Acceptance

The Hackathon is part of Generation AI Paris 2025 — a 3-day explosion of innovation with 8,000+ attendees, 300+ sessions, and 100+ speakers.

You’re getting the full package: keynotes, track sessions, workshops… and all the networking breakfasts and lunches.

Golden Pass Value: Your acceptance grants you a free pass to GenerationAI, apidays, AND GreenIO Paris—that’s 4 conferences normally valued at 999€, for the price of zero: To access the apidays conference component, use this promo link for a Regular Ticket.
We review applications daily — keep an eye on your inbox (and spam folder).
If your plans change, please release your spot so someone else can join.


🎟️ Ready?

Hit Register above.
Share your GitHub / LinkedIn / X handle and tell us what you’ll build.
Let’s connect LLMs and Graphs — and prototype the next leap in contextual intelligence, powered by Scalingo and Neo4j.

Ready for more?

Check out other posts from this blog.

View all posts

Contact Organizers

Questions? We're here to help.