Members-Only
Recent Talks & Demos are for members only
You must be an AI Tinkerers active member to view these talks and demos.
LangChain: Local LLM Orchestration
Compare building a multi-perspective LLM tool using pure Python, LangChain, and Prompt Flow to understand framework trade-offs on the same simple task.
Showing a prototype of how I am exploring LLM orchestration toolchains by building the same thing three ways: pure Python, LangChain, and Prompt Flow. The use case is a simple multi-perspective tool that sends a prompt through different ‘personality’ system prompts and captures the outputs. This is a learning spike, not a polished product—I’ll share what I built, what I learned, and where I got stuck.
Python repository demonstrating structured configuration and evaluation of LLM orchestration toolchains.
- LangChainThe open-source framework for building and deploying reliable, data-aware Large Language Model (LLM) applications.LangChain is the essential framework for engineering LLM-powered applications: it simplifies connecting models (like GPT-4 or Claude) to external data, computation, and APIs. The platform provides a modular set of components—Chains, Agents, Tools, and Memory—allowing developers to quickly build complex workflows like Retrieval-Augmented Generation (RAG) pipelines and sophisticated conversational agents. Its Python and JavaScript libraries, combined with LangChain Expression Language (LCEL), offer a standardized interface for rapid prototyping and moving applications to production with confidence.
- Prompt FlowPrompt Flow is your comprehensive tool for building reliable LLM-powered applications (like RAG systems or chatbots)Prompt Flow is your comprehensive tool for building reliable LLM-powered applications (like RAG systems or chatbots). It provides a visual, executable graph for orchestrating components: link Large Language Models (LLMs), custom Python code, and prompts together. Developers use it to easily debug flows, tune performance with prompt variants, and run large-scale testing (batch runs) using built-in evaluation flows. The platform integrates directly with Azure Machine Learning, providing enterprise-grade security and a smooth path to deploying real-time endpoints for your AI application backends.
- LM StudioLM Studio is the cross-platform desktop application for downloading, running, and serving local LLMs (Llama, Gemma, Qwen) on macOS, Windows, and Linux.LM Studio delivers local LLM power directly to your desktop (macOS, Windows, Linux). Use the built-in catalog to discover and download models like Llama 3.1 or Gemma 2 in formats like GGUF and MLX. Run these models offline via a simple chat interface, or expose them over an OpenAI-compatible REST API for seamless integration with your development projects. This is local, private AI: full control, zero cloud dependency.
- Qwen 3 4BThis is the Qwen 3 4B model: a highly efficient, 4This is the Qwen 3 4B model: a highly efficient, 4.0B parameter dense LLM from the Alibaba Cloud Qwen team. It's built for performance at scale, featuring the innovative hybrid 'Thinking' and 'Non-Thinking' modes for dynamic control over complex reasoning versus fast, general dialogue. The model supports a substantial native context length of 32,768 tokens, extendable to 131,072 with YaRN. Released under the permissive Apache 2.0 license, the Qwen 3 4B offers a powerful, compact solution for commercial deployment, rivaling much larger models on key benchmarks.
- PythonPython: The high-level, general-purpose language built for readability, powering everything from web backends to advanced machine learning models.Python is the high-level, general-purpose language prioritizing clear, readable syntax (via significant indentation), ensuring rapid development for any team . Its ecosystem is massive: use it for robust web development with frameworks like Django and Flask, or leverage its power in data science with libraries such as Pandas and NumPy . The Python Package Index (PyPI) provides thousands of community-contributed modules, offering immediate solutions for tasks from network programming to GUI creation . The language is actively maintained by the Python Software Foundation (PSF), with the stable release currently at Python 3.14.0 (as of November 2025) .
Related projects
Practical LLM Integration: Lessons from Building a SEC Filing Analysis Project
Houston
See a live coding demo building an LLM analysis system, focusing on cost reduction via regex, data extraction,…
Building a Product Powered by Local LLM's Only - Using LLM's in Cursor
Seattle
This talk covers building privacy-focused applications using locally hosted LLMs, challenges faced, creative solutions, and insights from a…
AI Security Overview and A Local LLM Walkthrough
Fort Wayne
Explore MITRE ATLAS and a local LLM chatbot demo using LangChain and Ollama to secure local data confidentiality…
Code Less, Create More: AI's productivity Revolution
Manchester Nh
Explore real-world AI tool use accelerating delivery, reducing debt, and shifting developers to innovation roles through practical examples.
Red teaming ideas: using LLM's to think through investment decisions
New York City
Explore how large language models can generate counter‑arguments to investment theses, moving beyond summarization, with a live demo…
How to Drink from the AI Firehose ...
San Diego
A concise AI-generated summary of podcast insights, framed for business leaders, delivering essential ideas in under a minute…