A few weeks ago, I wrote a post saying I had zero idea what “chunking” even was. Now, a few weeks later I definitely understand more, but not enough.That’s why today is fully dedicated to Pillar 1: Chunking. Back then I got great examples: 🍞 “Slice a loaf of bread into pieces.” 🍕 “Cut a pizza into slices.”Perfect analogies — and still true. But now I understand why chunking is so important: 🔹 What Chunking Really Is Chunking is the most critical preprocessing step in any RAG system. IT means breaking large documents into smaller, meaningful segments (“chunks”), which are then embedded, indexed, and retrieved later. Chunks are the atomic information units your RAG system uses. If the chunks are bad, retrieval is bad — and the LLM can’t fix it. 🔹 The Core Dilemma Chunking is always a balance between: 1️⃣ Precision – smaller chunks give cleaner embeddings 2️⃣ Context – bigger chunks give more meaning to the LLM Too big → diluted meaningToo small → missing context→ And THAT is the hardest challenge in chunking. 🔹 Best Practices for Chunking Here are the key strategies I’m learning: 📌 Recursive Character ChunkingRespects natural text boundaries (paragraphs, sentences).Often the recommended default. 📌 Overlap (10–20%) Ensures context isn’t lost at the edges.Example: 500-token chunk → 50–100-token overlap. 📌 Optimal Sizes A strong starting point is 512–1024 tokens per chunk. 📌 Advanced Methods– Semantic Chunking: uses embeddings to detect topic changes– Agentic Chunking: LLM splits text into atomic, meaningful statements These methods help avoid context loss and improve retrieval quality. 🔹 Why This Matters Chunking literally determines what your RAG system can find.And if retrieval fails, the LLM fails — it can’t magically invent the missing context. All resources, diagrams, and notes as always:👉 Notebook: https://notebooklm.google.com/notebook/ea1c87b2-0eda-43f8-a389-ba1f57e758ce

Really enjoyed hearing from the speakers and connecting with so many talented, like minded people in the automation community. Already planning follow-up meetings with some promising potential collaborations on the horizon. Events like these remind me why the tech community here is so special it's not just about the tools, it's about the people building with them. Big thanks to Bart Veldhuizen , Marrallisa Kreijkes and Tino Zwirs of n8n for organizing and making it happen! https://www.linkedin.com/feed/update/urn:li:activity:7402086121158955008/

CFO: "Reduce costs by 20%" R&D: "Without losing taste?" CFO: "Of course." This conversation usually ends in compromises. But what if it doesn't have to? Recent case: → Calories reduced from 783 to 600 kcal → Taste and aroma profile maintained → Costs decreased instead of increased → Compliance maintained How? AI optimization that analyzes thousands of combinations. Finds the sweet spot between all constraints. Cost, taste, nutrition, compliance. No more choosing. Just optimize. Curious and want to see a demo? DM me or leave a message below

As you perhaps know I'm helping students in Uganda with learning AI and building in n8n . They are on there way for 8 weeks now and I have Exciting news from our the team! I'm thrilled to share that everyone who submitted on time passed their second test an amazing achievement that demonstrates the incredible momentum this team is building! (very proud of them ) The Highlights: * 100% pass rate for on-time submissions * One team member achieved a perfect score, setting an outstanding benchmark for the group * Strong performances across the board, with scores ranging from 85-100 points * Clear evidence of growing technical skills AND professional communication abilities Key Growth Areas: Our team is mastering not just the technical execution (what is very important), but also the art of professional client communication. Clean email formatting, customer friendly messaging, and attention to detail are becoming second nature. These skills combined with technical expertise are what transform good work into exceptional, client ready deliverables. What's Next? The team is now embarking on an exciting new assignment (my favorite !!) : Developing AI business concepts. Each member will ideate, plan, and strategize around an AI powered business they'd like to launch. This isn't just theory it's about applying everything they've learned to real world entrepreneurial thinking. The Bigger Picture: Every test, every project, every piece of feedback is shaping these members into skilled, professional. The upward trajectory is undeniable, and I couldn't be more proud of their commitment to growth and excellence. Keep building. Keep learning. Keep pushing forward! #Uganda #AI #Innovation

80% of your results come from 20% of the effort. Today presented to 6 companies the AI food R&D platform . Result 4 clients want to make the next step But in traditional R&D, we spend: 80% of time on research and analysis 80% of time on trial-and-error 80% of time on routine work And only 20% on real innovation and optimization. Smart teams flip this: AI handles the first 80% in 20% of the time. Experts focus on the last 20% that makes the difference. From weeks to days. From exhaustion to energy. From frustration to flow. How would your team perform with this ratio? Do you want to work smarter let me know