Master Agentic AI with Abel Chin: Practical LLM Workshop

I’ve just published a deep‑dive on LangChain — one of the most exciting frameworks for building modular LLM applications. In this blog, I break down: 🔹 How prompts, chains, agents, and memory fit together 🔹 Why RunnableSequence is the new way to build workflows 🔹 How to use FakeListLLM + HuggingFace embeddings for offline demos 🔹 Real‑world use cases like customer support bots, research assistants, and workflow automation 💡 The blog includes runnable code snippets, a Jupyter Notebook template, and a GitHub‑ready README so you can try everything yourself. GitHub Link: https://lnkd.in/gGUD4f4K Medium Blog Link: https://lnkd.in/gjBXSexR LangChain isn’t just about chaining prompts — it’s about designing intelligent systems that combine reasoning, context, and external tools. #LangChain #LLM #AI #GenAI #Python #InnomaticsResearchLabs #LearningByDoing

If you're building AI apps with LangChain, knowing the right document loader can save a lot of time. Here’s a quick guide to common file types and their loaders: 📁 .txt → TextLoader 📄 .pdf → PyPDFLoader 📊 .csv → CSVLoader 🧾 .json → JSONLoader 📝 .md → UnstructuredMarkdownLoader 🌐 URLs → WebBaseLoader Using the right loader helps clean, structure, and prepare data for RAG systems, chatbots, and AI workflows. Save this post for later and follow for more AI content. #LangChain #Python #AI #MachineLearning #DataScience #LLM #RAG #Chatbot

🚀 Leveling up in Retrieval-Augmented Generation (RAG)! I’m currently diving into the core of modern AI development, focusing on: 1. RAG Architecture: Mastering practical, data-driven AI workflows. 2. LangChain & LlamaIndex: Building applications with the two leading Python ecosystems. 3. Gradio: Designing interactive, user-friendly interfaces for LLM apps. Turning static models into dynamic, knowledgeable assistants. 💡 Which tool is your favorite for building RAG: LangChain or LlamaIndex? #GenerativeAI #RAG #Python #LlamaIndex #LangChain #AI

🚀 Day 3 of Building My AI Code Review & Bug Prediction System Today was more about actually starting the implementation. 🔹 Set up the basic backend using Flask 🔹 Started working on API structure for code input and output 🔹 Began dataset collection for training the model 🔹 Explored how to preprocess code data for ML models 💡 Key Learning: Turning an idea into a working system is challenging — but breaking it into small steps makes it manageable. 📌 Next Step (Day 4): - Complete dataset preprocessing - Train a basic ML model - Connect backend with frontend Slowly turning this idea into reality 🚀 #AI #MachineLearning #Python #Flask #BackendDevelopment #StudentProject #LearningInPublic #ECE

If you're new to Agentic AI, don’t start with frameworks. Start with the code. Most people jump straight into LangChain, LangGraph, or AutoGen without understanding what an agent actually does. That’s like driving automatic without knowing manual. I took a different approach. Built a simple agent loop from scratch: No wrappers. Just Python. Here’s the idea: Step 1 - Plan Understand the query Decide the action Step 2 - Execute Call the tool Get real data Step 3 - Answer Use LLM with that data Generate grounded output No data? Say it clearly. That’s it. This is what most frameworks do just with more layers. Learn this once, and frameworks become optional. If you want to learn Agentic AI: Build it yourself first. 🔗 Repo: [https://lnkd.in/dMwdynU2] #AI #Rag #AgenticAI #LLM #Python #BuildInPublic

🚀 Day 17 of My Generative & Agentic AI Journey! Today’s focus was on Passing Arguments to Functions and understanding how different data types behave. Here’s what I learned: 🔤 Passing Strings vs Lists: • Strings are immutable → cannot be changed inside a function 👉 Any modification creates a new value • Lists are mutable → can be updated inside a function 👉 Changes reflect outside the function as well 📥 Types of Arguments: • Positional Arguments → Values are passed in order 👉 Example: first value goes to first parameter, second to second • Keyword Arguments → Values are passed using parameter names 👉 Order doesn’t matter here ⚙️ *args and **kwargs: • *args → Used to pass multiple positional arguments 👉 Treated as a tuple inside the function • **kwargs → Used to pass multiple keyword arguments 👉 Treated as a dictionary inside the function 💡 Key takeaway: Understanding how arguments work helps in writing flexible and reusable functions. Getting more comfortable with writing dynamic and scalable Python code 🚀 #Day17 #Python #GenerativeAI #AgenticAI #LearningJourney #BuildInPublic

Generate datasets and use them immediately without breaking your workflow. With the DataCreator AI SDK, you can: • Generate conversational datasets directly in Python • Save as JSONL (fine-tuning ready) • Train your model in the same notebook No switching tools. No manual data prep. You can go from: idea → dataset → training in one flow. That’s how it should be. If you’re working with LLMs, try integrating this into your pipeline and see if it actually saves you time. Also, if something breaks or feels off, please add it under issues on our GitHub repository. Feedback (and bugs) are welcome. Repo + details in the comments 👇 #ai #sdk #generativeai #syntheticdata

We are on GitHub! Follow, Star, and Fork the DataCreator AI SDK and generate conversational datasets quickly, in the same notebook where your training code resides. #ai #generativeai #github

23,602 stars and countingOpenAI's new 'openai-agents-python' framework exploded by 752 stars today, and it's only been live for a week. This lightweight framework is pure Python, tailored for building multi-agent workflows. Unlike bloated architectures, it keeps things lean while enabling complex, coordinated tasks across agentic AI systems. Whether you're working on distributed decision-making or collaborative LLMs, the design is meant to get out of your way and let you prototype fast. This feels like OpenAI signaling more focus on scalable, accessible multi-agent AI. Multi-agent systems aren't just a cool ideatheyre a necessary evolution for AI to tackle real-world challenges where single-agent models hit their limits. Expect to see this framework pop up in research papers and production stacks soon. Whats the first multi-agent workflow you'd try to build with it? Read more: https://lnkd.in/d4_jBcCx #AIAgents #LLM #PythonFramework #MultiAgentSystems #OpenSourceAI

Here’s my AI chatbot in action — handling real-time, multi-turn conversations while keeping full context. Built with Flask + Meta’s Llama 4 Maverick (NVIDIA NIM), and powered by a memory system that makes interactions feel natural and coherent. 💬 From user input → API → model → response → stored memory — all working together seamlessly. 💻 Check it out on GitHub: https://lnkd.in/dSg6mX4R Curious to hear your thoughts 👇 #AI #Chatbot #Python #Flask #MachineLearning #FullStack #LLM