Speaking

Sharing insights on AI engineering, technical leadership, and system design

8+ Talks Given
2 Countries
2 Podcasts
  • New GenAI:
    Industrialiser sans compromettre

    Guide terrain pour accélérer sa delivery GenAI - Stratégie hybride pour accélérer son time-to-market et préserver sa différenciation. Analyse des trade-offs build/buy pour chaque composant critique, critères de décision et métriques de succès. Patterns d'architecture issus de déploiements réussis en contexte contraint.

    AI Strategy GenAI
  • dos and don'ts
    developing using genAI tools

    Sharing practical insights on effectively integrating GenAI tools into software development workflows. Drawing from 18 months of hands-on experience, this talk explores essential tools, enhanced development cycles with AI, and proven patterns to maximize productivity while avoiding recurring issues of GenAI spaghetti code mess. Learn how to craft high-quality prompts, leverage AI for the complete feature creation workflow - ideation, testing, and code reviews.

    Ignite talk Developer tool GenAI
  • Generative AI for Organizations: Myths, Limits and Opportunities

    Moderating a roundtable at AGIR 2024, organized by the Gendarmerie Nationale, discussing the concrete challenges of LLM industrialization, from infrastructure choices to adoption strategies and governance.

    LLM Enterprise AI Strategy
  • GenAI in prod: practical tech & product insight for delivering value with RAG

    Drawing from extensive interviews with orgs at various stages of their GenAI journey, this talk delivers practical insights into building effective RAG solutions. We'll explore i) technical challenges—from model selection to performance evaluation—and ii) product strategy considerations that make or break RAG implementations. Whether you're a technical leader or ML engineer, you'll gain actionable insights from real-world deployments, to help you build your unique value prop using genAI in your business vertical.

    RAG Product GenAI Unique Value Proposition Production
  • LLM Finetuning: From AGI Promises to Production Realities

    From AGI promises to real-world implementation challenges, this talk cuts through the hype to deliver practical insights on LLM fine-tuning for enterprise applications. Drawing from both startup and corporate deployments, we'll explore effective PEFT methods, reinforcement learning approaches, and evaluation frameworks—equipping you with the knowledge to transform off the shelves LLMs into domain-specific tools. We'll focus on actionable insights that will help you build a successful roadmap for your organization's LLM deployment.

    LLM Fine-tuning ML Ops LoRA PEFT NLP Quantization
  • Jeune Investisseur Podcast

    Producer and co-host of the Jeune Investisseur podcast. Jeune Investisseur is a 4-episode mini-series on personal finance and investing. From how to manage your budget to setting up and monitoring an investment strategy, Vincent and François share their protips for achieving your financial goals. Not financial advice. Do your own research.

    Personal finance Investing Financial literacy
  • Post Mortem Podcast

    Producer and host of the Post Mortem podcast. In Post Mortem, engineers reflect on real-life incidents of IT systems they experienced. In each episode, we zoom on a specific event, ranging from a system outage, a cyber-attack, or a machine learning algorithm going wild with production data. We try to understand what happened and how the people behind those systems solved the situation. Along the way, you'll get hands-on advice shared by experienced practitioners that you can implement within your team to limit the risk of such incidents.

    Engineering Incidents Leadership Interview
  • Software has already eaten the world; now what? A case for open source software

    In this talk, I advocate for a deeper commitment to open source software (OSS) by sharing both successes and challenges when integrating OSS tools into real-world projects. By fostering transparency and collaboration, developers not only strengthen the OSS ecosystem but also equip the community to make informed choices about the best tools for their needs. I also explore how OSS serves as a critical lens for evaluating proprietary software, empowering organizations to discern between true innovation and repackaged features. This talk is a call to action for developers to contribute authentically to OSS, creating a culture of trust, accountability, and shared growth in software development.

    Open Source TensorFlow Enterprise