Tool Use - AI Conversations

Join the Tool Use Discord: https://discord.gg/PnEGyXpjaX

Vector databases can be an important component for building reliable AI agents and scalable semantic search applications. In this episode, Arjun Patel from Pinecone breaks down how to optimize your RAG pipeline, choose the right embedding models (sparse vs. dense), and implement effective chunking strategies for better data retrieval. We also explore the new Pinecone plugin for Claude Code, demonstrating how to build a recommendation system and chat with your documents using Pinecone Assistant without writing complex code.

https://www.pinecone.io/
https://www.linkedin.com/in/arjunkirtipatel/

Connect with us 
https://x.com/ToolUsePodcast
https://x.com/MikeBirdTech 

00:00:00 - Intro
00:01:11 - What Vector Databases Unlock
00:04:40 - Optimal Chunking Strategies for RAG
00:09:07 - How Embedding Models Work
00:17:25 - Improving Search with Re-ranking
00:26:52 - SQL vs Vector Database Architecture
00:35:48 - Claude Code & Pinecone Assistant Demo

Subscribe for more insights on AI tools, productivity, and vector databases.

Tool Use is a weekly conversation with the top AI experts.

Join the Tool Use Discord: https://discord.gg/PnEGyXpjaX

Unlock the power of AI fine-tuning for image and video models with Greg Schoeninger, CEO of Oxen.ai. In this episode, we explore how to move beyond simple prompt engineering to training custom open-source models for as little as $1. Discover the technical strategies for building high-quality datasets, the trade-offs between LoRAs and full fine-tunes, and how to achieve consistent characters and styles in generative video.

We dive into real-world examples, including how to generate massive product catalogs for a fraction of the cost of enterprise APIs and the story behind the viral "Isometric NYC" project built with Claude Code. Greg breaks down the entire AI lifecycle—from data curation and labeling with Vision Language Models (VLMs) like Qwen and Gemini to deploying efficient, specialized models that outperform general-purpose giants. Whether you are looking to optimize GPU costs, automate video workflows, or build your own AI tools, this conversation provides the blueprint for scaling your AI capabilities.

Links from the episode:
Oxen AI: https://www.oxen.ai/
Fine Tuning Fridays: https://luma.com/oxen
Isometric NYC Project: https://cannoneyed.com/projects/isometric-nyc

Connect with us 
https://x.com/ToolUsePodcast
https://x.com/MikeBirdTech 
https://x.com/gregschoeninger

00:00:00 - Intro
00:01:47 - When to Switch from Prompt Engineering to Fine-Tuning
00:11:05 - How to Build a Dataset for AI Fine-Tuning
00:21:52 - LoRA vs. Full Fine-Tuning Explained
00:27:53 - Case Study: Fine-Tuned Qwen 3 VL vs. Google Gemini
00:36:33 - Case Study: Isometric NYC with Claude Code

Subscribe for more insights on AI tools, productivity, and fine-tuning.

Tool Use is a weekly conversation with the top AI experts.

Join the Tool Use Discord: https://discord.gg/PnEGyXpjaX

Unlock the ability to build AI tools without coding experience as Justin Borge joins Tool Use to demonstrate how he built Helios, a personal management system, using only LLMs like Gemini and Claude. Justin breaks down his transition from marketer to growth engineer, sharing his exact workflows for using JSON documentation to give AI a perfect memory of his business and goals. He reveals his unique "bullying" strategy for debugging code using persona-based prompts and explains how he created a human-in-the-loop system to safely edit local files without hallucinations. This conversation covers everything from writing your first Python script to mastering meta-prompting for maximum productivity.

https://www.linkedin.com/in/justinborge/
https://justinborge.com

Connect with us
https://x.com/ToolUsePodcast
https://x.com/MikeBirdTech

00:00:00 - Intro
00:03:49 - Top AI Tools & JSON Workflows
00:10:06 - Transitioning from Marketing to Engineering
00:12:23 - How to Debug Code Without Coding Knowledge
00:13:56 - Persona Prompting & "Bullying" LLMs
00:18:09 - Live Demo: Helios AI Management System
00:31:29 - How to Start Building AI Tools Today

Subscribe for more insights on AI tools, productivity, and growth engineering.

Tool Use is a weekly conversation with the top AI experts.

Join the Tool Use Discord: https://discord.gg/PnEGyXpjaX

Discover the power of the Ralph Wiggum Loop with Ryan Carson as we explore the future of AI coding agents, autonomous workflows, and building one-person companies. In this episode, we break down how to implement Jeff Huntley’s simplified bash script loop using tools like Claude Code and Amp to ship features faster than ever. Ryan shares his tactical advice on creating high-quality PRDs, defining clear acceptance criteria for agent browser testing, and moving beyond human-in-the-loop development. We also dive into advanced AI engineering strategies, including self-healing systems, MCP servers like Sentry and Fizzy, and how to use LLMs to automate business operations through CRON job insights and email marketing drips. Whether you are an AI engineer or a founder, learn how to stop writing syntax and start orchestrating a fleet of agents to amplify your delivery speed and creative joy.

Follow Ryan's Work:
Untangle: https://www.untangle-us.com/
Ralph Loop Repo: https://github.com/snarktank/ralph
Ryan on X: https://x.com/ryancarson

Connect with us
https://x.com/ToolUsePodcast
https://x.com/MikeBirdTech
https://x.com/ryancarson

00:00:00 - Intro
00:01:17 - How the Ralph Wiggum Loop Works
00:02:46 - Fresh Context Windows vs. Context Rot
00:12:15 - Deterministic Harnessing and Agent Hooks
00:17:59 - Agent Browser Testing & Acceptance Criteria
00:28:13 - Automating Business and Marketing Loops
00:39:09 - PR Review Bots and AI Engineering Workflows

Subscribe for more insights on AI tools, productivity, and orchestration.

Tool Use is a weekly conversation with the top AI experts.

Join the Tool Use Discord: https://discord.gg/PnEGyXpjaX

Claude Code is evolving from a terminal-based developer tool into the Everything Agent that manages your entire life. In this episode, Eric Buess returns to show us how he built a hands-free, voice-controlled AI system that handles everything from school emails and calendar scheduling to monitoring family text messages. We dive deep into advanced Claude Code strategies, including how to use the hook system to intercept model prompts and the Ralph Wickham loop to enable truly autonomous task completion. Eric explains why he moved away from static claude.md files in favor of programmatic hooks that steer the model with surgical precision. We also explore the power of adversarial sub-agents that fact-check AI claims and the Titus wake word system that lets him command his terminal from a car or Meta Ray-Ban glasses. Whether you are a developer looking to maximize the Anthropic Agent SDK or a productivity enthusiast wanting to reclaim your time, this masterclass in agentic workflows will change how you think about AI automation.

Keep up with Eric: https://ericbuess.ai

Connect with us
https://x.com/ToolUsePodcast
https://x.com/MikeBirdTech
https://x.com/EricBuess

00:00:00 - Intro
00:02:20 - The Claude Code Hook System & Vertical Stack
00:03:08 - Why Agentic Hooks Matter for Automation
00:11:58 - Replacing claude.md with JSON Rules & Hooks
00:16:43 - Multi-Agent Workflows & Adversarial Validation
00:29:58 - Autonomous Task Loops & Self-Referential Injection
00:41:00 - Hands-Free AI: Voice-Controlled Claude Code

Subscribe for more insights on AI tools, productivity, and Agentic Workflows.

Tool Use is a weekly conversation with the top AI experts.