GenAI Product Maker ⚡️ Cofounder & CTO of Wikit 🤖
Lyon, Rhône-Alpes
Joined October 2010
- Tweets426
- Following219
- Followers120
- Likes466
Christophe Bouvard retweeted
Memory in AI agents seems like a logical next step after RAG evolved to agentic RAG.
RAG: one-shot read-only
Agentic RAG: read-only via tool calls
Memory in AI agents: read-and-write via tool calls
Obviously, it's a little more complex than this.
I make my case here: leoniemonigatti.com/blog/fro…
Christophe Bouvard retweeted
Context engineering is a far better term than prompt engineering or RAG.
It's helpful to think of it in 2 stages
1. Gather - get all possible relevant information
2. Glean - cull down only the information needed
Christophe Bouvard retweeted
A great talk at the AI Engineer World's Fair! @chu_onthis, PM of MCP at @AnthropicAI, shared insights unveiling MCP's actual value:
- The origin story of MCP
- Why Anthropic have launched it
- What real opportunities it gives
Here are the key points:
Christophe Bouvard retweeted
A free guide from @AnthropicAI "How we built our multi-agent research system"
Explains:
- How their Research system's architecture works
- Prompting and testing
- Production challenges
- Benefits of multi-agent system
Really worth reading: anthropic.com/engineering/bu…
Christophe Bouvard retweeted
If you're building AI agents, read this twice:
There’s a new invisible OS being written for the autonomous future. It’s built on two emerging protocols:
• Model Context Protocol (MCP)
• Agent-to-Agent Protocol (A2A)
Here’s why they’re a big deal—and why you need to care:
Christophe Bouvard retweeted
Nomic Embed Text V2 is now available
- First general purpose Mixture-of-Experts (MoE) embedding model
- SOTA performance on the multilingual MIRACL benchmark for its size
- Support for 100+ languages
- Truly open source - open training data, weights, & code
- Apache 2.0 License
Christophe Bouvard retweeted
Our newest project is taking flight: meet codename goose! 🪶
Today, we launched an open source on-machine AI Agent. It’s modular, works with your preferred LLM, and integrates seamlessly with developer tools and other software via MCP.
block.github.io/goose/blog/2…
Christophe Bouvard retweeted
A free comprehensive book on Foundations of LLMs
In this 230-page book, you can learn everything you need to know about:
- Pre-training (with examples)
- Generative LLMs
- Their training and scaling laws
- All about prompts and prompting methods
- Instructions to LLMs alignment
Check it out here: arxiv.org/abs/2501.09223
Christophe Bouvard retweeted
Google recently published this great whitepaper on Agents.
2025 is going to be a huge year for AI Agents.
Here's what's included:
- Introduction to AI Agents
- The role of tools in Agents
- Enhancing model performance with targeted learning
- Quick start to Agents with LangChain
- Production applications with Vertex AI Agents
Great place to start learning about AI Agents.
Christophe Bouvard retweeted
I'll get straight to the point.
We trained 2 new models. Like BERT, but modern. ModernBERT.
Not some hypey GenAI thing, but a proper workhorse model, for retrieval, classification, etc. Real practical stuff.
It's much faster, more accurate, longer context, and more useful. 🧵
Christophe Bouvard retweeted
In this AI Agents Masterclass with @arizeai learn from our founder @jerryjliu0 how to build a knowledge-backed agent using LlamaIndex workflows!
Key takeaways:
⭐️ Basic agent pattern = LLM router + tools
⭐️ Event-based vs. graph-based architecture
⭐️ LLM routers preferred over simple classifiers for better generalization
⭐️ State management critical: local state via events, global state for complex data
⭐️ Debugging/observability achieved through traces and step inspection
⭐️ Function calling accuracy not 100% - needs evaluation and prompt improvements
⭐️ Start simple before adding complexity like global state or persistence
youtube.com/watch?v=UzB3lMw_…
Christophe Bouvard retweeted
🥳 The LLM Engineer's Handbook is officially released!
I'm really proud of this book.
It's a comprehensive guide to building and deploying LLM systems in the real world.
This is the culmination of @iusztinpaul's and my experience writing LLM courses and articles online.
Christophe Bouvard retweeted
Mozilla's new blog post w/ executive summary on some key progress we've made on @llamafile the past several months. hacks.mozilla.org/2024/10/ll…
Christophe Bouvard retweeted
Not all retrieval metrics are created equal.
Precision: How many retrieved items are relevant?
Recall: How many relevant items were retrieved?
MRR: How relevant is the top hit?
MAP: How is the average precision across all queries?
NDCG: How well-ordered are the results?
Measure what matters. Choose the right metric.
Christophe Bouvard retweeted
Large Language Model Agents is the next frontier. Really excited to announce our Berkeley course on LLM Agents, also available for anyone to join as a MOOC, starting Sep 9 (Mon) 3pm PT! 📢
Sign up & join us: llmagents-learning.org
