Sébastien Brochet

Principal Consultant @ Microsoft - AZ-104/AZ-204/AZ-400/AI-900 - #azure #devops #genAI

[Sébastien Brochet @sebbrochet]

♟️ Ch3ssVid5 — sync chess games with YouTube video commentary. The board follows the video automatically.

Free, open-source, no account needed.

🔗 https://t.co/G72Rp1a3dB

[Sébastien Brochet @sebbrochet]

📚 Ch3ssVid5 Hub has 1000+ #chess videos from GothamChess, Blitzstream, Hanging Pawns & more —

What's missing? Moves and timestamps. That's where you come in.

Pick a video, add moves/timestamps, submit a PR.

https://t.co/HXpYXOxv5I

[Sébastien Brochet @sebbrochet]

En route pour l'#AI Tour Paris by #microsoft à la Porte de Versailles https://t.co/41JhbmzxUC Répondez #hello si vous y serez également ! :)

[Sébastien Brochet @sebbrochet]

@IDFmobilites Je vous confirme que les tourniquets à #Invalides (Metro vers RER C) sont buggés et décomptent un nouveau ticket alors que le ticket actuel est toujours valide ! Je me suis fait avoir plusieurs fois. Comment se faire rembourser ? Merci

[Sébastien Brochet @sebbrochet]

The AI handled drafting.
I handled intent.
That’s the real shift:
👉 AI doesn’t replace expertise — it amplifies it.

[Sébastien Brochet @sebbrochet]

I ran a small experiment: vibe‑authoring a technical e‑book with AI.
Over a few weeks, I created “Copilot Studio Gems 💎” and published it as a website:
👉 https://t.co/ldqQkVK7wZ

#CopilotStudio #GenerativeAI

[Sébastien Brochet @sebbrochet]

I wasn’t really a “writer”.

My role was closer to editor/curator/architect:
. Shaping concepts.
. Selecting what matters.
. Refining quality.
. Building the site.

[Sébastien Brochet @sebbrochet]

C'est parti pour une nouvelle saison où j'ai le plaisir d'accompagner la #startup https://t.co/WAKlrAKug2

[Microsoft France @microsoftfrance]

Avec la 3e cohorte du Microsoft GenAI Studio, nous continuons à soutenir l’émergence des futurs champions français de l’IA générative 🤝
Une nouvelle promotion de 15 startups démarre un accompagnement intensif aux côtés de nos experts et partenaires.
👏 Félicitations aux lauréats !
https://t.co/sRu4QcKiVG

[Sébastien Brochet @sebbrochet]

Nice one slide summary, let's not forget #Microsoft Agent Framework (https://t.co/obfHUIMHkj) in the mix!

[Tech with Mak @techNmak]

AI Agents Cheat Sheet
[ Read. Bookmark. Repost ]

------------

This is a good starting point if you're trying to make sense of AI agents.

There’s a lot of talk about agent frameworks right now, but at the core, most of them build on the same set of ideas.

This cheat sheet gives a simple overview of the key building blocks, from LLMs to orchestration to protocols. Useful whether you’re exploring agent tooling, building internal automations, or just trying to understand the space better.

What is an AI agent?
- Agents combine reasoning with the ability to take action.
- They don’t just respond, they can plan, call tools, access data, and trigger real-world effects.

Language Model
- This is the core reasoning engine. It interprets input and generates plans or responses.
- But by itself, it can’t take real-world actions.

Tools
- APIs, functions, and external integrations that agents use to do useful work, like querying a database, sending an email, or calling a webhook.

Orchestration Layer
- This layer coordinates what the agent does, how it reasons (via CoT, ReAct, etc.), how it sequences steps, and how it interacts with tools.

Agentic Protocols
- Protocols like MCP and A2A enable agents to collaborate across platforms (e.g., Slack, GitHub) and maintain context across tasks.

Building AI Agents
- There’s no single way to build an agent.
- Some start with a single prompt. Others use low-code platforms. Some teams build full custom frameworks.

The cheat sheet maps out the trade-offs.

If you’re trying to understand the agent space, or explain it to your team, this breakdown might help.

Save it if you want a reference to come back to.

-----------

Follow @techNmak for regular insights.

[Sébastien Brochet @sebbrochet]

C'est bien noté et j'y serai ! #GenAI #microsoft #Startups

[Microsoft France @microsoftfrance]

La mauvaise nouvelle : C'est déjà la fin des candidatures pour la 3e édition du Microsoft GenAi Studio.
La bonne nouvelle : vous avez été +180 startups françaises à tenter votre chance !
RDV le 5 janvier pour découvrir les lauréats !

[Sébastien Brochet @sebbrochet]

Interesting, needs to dig this deeper...

[Avi Chawla @_avichawla]

DevOps vs. MLOps vs. LLMOps:

Many teams are trying to apply DevOps practices to LLM apps.

But DevOps, MLOps, and LLMOps solve fundamentally different problems.

DevOps is software-centric. You write code, test it, and deploy it. The feedback loop is straightforward: Does the code work or not?

MLOps is model-centric. Here, you're dealing with data drift, model decay, and continuous retraining. The code might be fine, but the model's performance can degrade over time because the world changes.

LLMOps is foundation-model-centric. Here, you're typically not training models from scratch. Instead, you're selecting foundation models and then optimizing through three common paths:

- Prompt Engineering
- Context/RAG Setup
- Fine-Tuning

But here's what really separates LLMOps: The monitoring is completely different.

In MLOps, you track data drift, model decay, and accuracy.

In LLMOps, you're watching for:

- Hallucination detection
- Bias and toxicity
- Token usage and cost
- Human feedback loops

This is because you can't just check if the output is "correct." You need to ensure it's safe, grounded, and cost-effective.

The evaluation loop in LLMOps also feeds back into all three optimization paths simultaneously. Failed evals might mean you need better prompts, richer context, OR fine-tuning.

So it's not a linear pipeline anymore.

One more thing: prompt versioning and RAG pipelines are now first-class citizens in LLMOps, just like data versioning became essential in MLOps.

And the ops layer you choose should match the system you're building.

👉 Over to you: What does your LLM monitoring stack look like right now?
____
Find me → @_avichawla
Every day, I share tutorials and insights on DS, ML, LLMs, and RAGs.

[Sébastien Brochet @sebbrochet]

Excellent!

[Sébastien Brochet @sebbrochet]

@laboite_a_pizza Merci pour eux et merci pour nous !

[Caroline Brochet @BrochetCaroline]

Merci La Boîte à Pizza de venir offrir des pizzas aux équipes de garde. Chacun à sa manière, chacun à son niveau, on participe tous à la lutte contre le coronavirus. La victoire contre le covid-19 sera celle de toute la Nation!

[Caroline Brochet @BrochetCaroline]

L’Association Professionnelle de Sages-Femmes et La Boîte Rose lancent une opération d’urgence pour soutenir les soignants et les accouchées Covid+. Protégeons les femmes enceintes et les nouveau-nés. Ils sont la Vie de demain. Participez et retweetez !
https://t.co/6yUZdLe8nO

[Caroline Brochet @BrochetCaroline]

Les seuls autorisés à sortir, ce sont eux! Ils sont notre espoir, notre force, notre futur. Protégez la Vie, restez chez vous! #RestezChezVous

[Caroline Brochet @BrochetCaroline]

AP-HP/Hôpital Trousseau. Lorsque le coronavirus vole des vies, les femmes continuent à donner la Vie. Les sages-femmes sont là pour elles. Fière d’être au service des femmes, au service de la Nation, au service de la Vie. #covid-19

[Sébastien Brochet @sebbrochet]

Just got #AZ400 certification (#devops expert) yesterday! :) https://t.co/cF35wmQfRy

[Sébastien Brochet @sebbrochet]

Just finished "Understanding #Enterprise #Architecture" course from #Pluralsight. It gives a good overview of the domain and associated topics, with a dedicated part to align with #Agile principles: https://t.co/O3m5qe34WX

[Sébastien Brochet @sebbrochet]

Been tuning my internet #VDSL connection. Disconnected unused home connectors that were attenuated signal, kept only the one to internet box: 20/1 to 37/7 (D/U Mbps)

[Sébastien Brochet @sebbrochet]

#Scrum PSM I exam passed! 🙂 Thanks #Microsoft for the training & exam. Eager to apply this framework with my customers while driving their #DigitalTransformation...