Agentic AI: Moving from a single prompt to 15 agents (AI Malt Academy)
Agentic AI: Moving from a single prompt to 15 agents In an era where GenAI is moving from simple prompts to autonomous systems, the challenge for organizations is turning scattered information into structured, decision-ready data. To address this crucial topic, we are delighted to welcome Thibault Boutet, a freelance AI Engineer specializing in production-grade AI agents with international experience across Europe and the United States. Formerly at Siemens Healthineers and currently collaborating with companies like Barnacle Labs, Thibault designs and builds complex multi-agent systems, particularly for demanding sectors like the pharmaceutical domain. In this webinar, Thibault shares his expertise on moving Beyond the Single Prompt. He explores how to design a multi-agent system where 15 specialized sub-agents work together to read, extract, and write structured sections. He breaks down the "Ops" side of GenAI—from implementing a crucial verification layer to ensure reliability, to mastering cost predictability and traceability. Whether you are a freelancer helping clients automate their complex workflows or a tech lead aiming to build production-ready AI architectures, this video provides the essential keys to scaling Agentic AI with confidence. Thank you to Thibault for sharing his expertise! Stay tuned for more insightful sessions from Malt Academy. Video Chapters 0:00 Introduction & Malt Academy welcome 2:15 Who is Thibault Boutet? From Siemens to AI Engineering 5:30 The Shift: Moving beyond the "Single Prompt" limitation 8:45 Defining Agentic AI: Autonomy vs. Linear Automation 12:20 Architecture Overview: Designing a 15-agent system 16:05 Specialized Sub-agents: Reading, Extraction, and Synthesis 20:30 The "Ops" of GenAI: Managing reliability and consistency 24:15 The Verification Layer: How to ensure agents don't hallucinate 28:50 Cost Predictability: Scaling agents without breaking the bank 33:10 Traceability: Auditing the decision-making process of an agent 37:40 Real-world Case Study: Multi-agent systems in the Pharma sector 42:15 Tooling & Stack: Frameworks for building production-ready agents 46:30 Future Outlook: The evolution of autonomous workflows 49:10 Q&A: Handling latency and complex prompt chaining ______________________________________________________________ Find all sessions covering various themes (tech, data, marketing, etc.) hosted by freelancers, clients, and Malt team members at https://www.malt.fr/resources/events 📚 Subscribe to our channel here: @maltfrance 🚀 Ready to transform the way you work? Sign up on Malt : https://www.malt.fr/signin 🗞️ Malt Newsroom: https://newsroom.malt.com/fr 💌 Our monthly Newsletter: https://www.malt.fr/c/newsletter 🥳 Our Events Newsletter: https://www.malt.fr/c/newsletter-event 👩🏼💻 Our LinkedIn: / maltcommunity

Andrej Karpathy: From Vibe Coding to Agentic Engineering w/ Stephanie Zhan

Building AI Agents that actually work (Full Course)

Build High-Performance AI Pipelines With Real-Time Streaming And Cortex Code

Freelancing with AI: Seizing Opportunities, Mastering the 5 Top Risks

Full Workshop: Build Your Own Deep Research Agents - Louis-François Bouchard, Paul Iusztin, Samridhi

Don't learn AI Agents without Learning these Fundamentals

AI Malt Academy : "AI what is different and why security matters" 🤖🛡️

The AI Tool Starter Kit: ChatGPT, Notion AI, Perplexity and Beyond

AI Agents Full Course 2026: Master Agentic AI (2 Hours)

Leading in the Age of AI: A Conversation with NVIDIA CEO Jensen Huang | Global Conference 2026

From Idea to $650M Exit: Lessons in Building AI Startups

E-invoicing, the final countdown: get compliant

Why AI Agents fail in the real world and how to make them production read (AI Malt Academy)

RL for Agents Workshop - Deep Dive on Training Agents with RL and Open Source

How AI agents & Claude skills work (Clearly Explained)

CLAUDE CODE ADVANCED FULL COURSE (3 HOURS)

What AI Agent Skills Are and How They Work

Full Archon Guide - Build AI Coding Harnesses That Actually Ship (LIVE)

Stop prompting, start building real solutions with AI

