☕ Builders & Brews: A Day of Agents, Coffee, and Code

On July 7th, 2026, AI Tinkerers Paris is hosting a hands-on builder day focused on real-world, web-grounded agents. This is a collaborative, zero-pitch co-working space designed for engineers, researchers, and technical founders who want to push the boundaries of autonomous workflows.

We are moving beyond simple chat prompts. This event is about building agents that interact dynamically with the live web, orchestrate complex tool-calling, and operate reliably. Whether you are optimizing production agentic loops or experimenting with new LLM infrastructure, you will find active builders in the room solving the same technical challenges.

Registration is required. Because space is limited and our events routinely exceed capacity, all applicants are screened based on technical merit and active builder status.

What You Get

• ⚡ Tavily Credits: Access to purpose-built web search and content extraction APIs optimized for LLMs and RAG pipelines.
• ⚡ Nebius (TBC) Token Factory Credits: High-performance LLM infrastructure and compute tokens to power your agent models.
• ⚡ Mulerun Access: Free access to The Always‑On AI Agent that Automates Your Workflows End‑to‑End.
• 🧱 Time & Space to Build: A dedicated co-working environment to start a new project or improve an existing workflow.
• 💬 Direct Engineering Access: Sit down and debug directly with engineers from the Tavily and Nebius teams.
• ☕ Premium Fuel: Free high-quality coffee and decaf options throughout the entire day.

Who Should Attend

This builder day is curated strictly for hands-on practitioners. It is designed for:

• Mid-Level & Senior Developers: Looking to bridge the gap between autocomplete APIs and fully autonomous, web-grounded workflows.
• AI Founders & CTOs: Building lean engineering teams that leverage agentic infrastructure to operate with massive leverage.
• Machine Learning Researchers: Experimenting with real-time web data ingestion, tool-calling evaluation, and local model orchestration.

Event Schedule

☕ 10:00 AM: Doors open
Come by early, settle in, co-work, and grab coffee on us.

🎤 10:30 AM: Partner intros
Quick intros from our amazing partners.

💻 Afterwards: Build & collaborate
Spend the day building, working, and experimenting with free credits from partners.

🎶 Around 4:00 PM: DJ vibes
We’ll wind down the day with some chill music and good energy, definitely stick around for it.

The Stack: Web-Grounding & LLM Infrastructure

To help you ship working code during the session, our partners are providing direct support for their developer suites:

• Tavily: Learn how to leverage real-time web queries that crawl up to 20 sites simultaneously, returning pre-scraped, pre-ranked text chunks under 500 characters directly to your LLM context window.
• Nebius: Deploy models on sovereign, high-efficiency GPU infrastructure and leverage optimized token generation pipelines.
• MuleRun: Set up proactive monitoring for the metrics that matter: website uptime, data pipeline health, competitor pricing, inventory levels, campaign performance.

AI Tinkerers Paris Stats

• Attendees: AI Tinkerers - Paris connects 4,263 elite technical professionals, with over 65% specializing in generative AI, RAG, and LLM fine-tuning, alongside 42% skilled in MLOps and cloud infrastructure. The community is distinguished by a high concentration of active open-source contributors and technical founders from leading organizations like Hugging Face, Mistral AI, and top-tier French research institutions.
• Companies Represented: Featuring tech giants like Google, Meta, Microsoft, and Amazon, leading AI pioneers such as Hugging Face, Mistral AI, and ElevenLabs, alongside high-growth startups and platforms like Datadog, Docker, Algolia, Chainlit, and more
• Demos: 131 demos have been submitted and 84 have been presented. The most exciting themes have centered on agentic AI moving into production, knowledge-graph-enhanced retrieval and structured generation, and reliability via evaluation and simulations. Technical exploration has frequently covered orchestration/state-machine patterns, graph-structured context, infrastructure-native deployment (serverless/containers), multimodal/UI integrations, and efficiency gains through latency/cost and on-device execution.
• Testimonials:

  “…Strong demo with a clear builder takeaway. Turning massive financial news flow into structured, machine-readable signals made the value of NewsWitch immediately understandable…”

  — AI consultant, NLP

  “…The research-to-portfolio perspective, especially around macro data, ML decision inputs, Ex Machina, and internal LLM/agent usage, gave builders a real operating view…”

  — AI consultant, NLP

  “…This is definitly the subject of the year!…”

  — AI consultant, NLP

  “…Excellent technical depth. The Temporal Fusion Transformer approach to futures microstructure was compelling…”

  — AI consultant, NLP

  “…Having so many volunteers involved from the community was exceptional…”

  — AI consultant, NLP

  “…Thanks first and foremost to the mentors and volunteers!…”

  — AI consultant, NLP

  “…Great job! I enjoyed my first event very much and I’m impressed by the quality of the quorum…”

  — Finance @ AI startup

Access & Logistics

Registration is open now. Accepted builders will receive a confirmation email containing the exact venue details in Paris. Please ensure your community profile is complete with links to your GitHub, LinkedIn, or technical portfolio to facilitate the screening process.

Sponsor the builders shaping what’s next