AI Inference
549 episodes · Page 9 of 23
#4919: Parakeet vs Whisper: On-Device Dictation Showdown
NVIDIA Parakeet beats Whisper on phone dictation despite having more parameters. Here's why architecture matters more than model size.
#4836: When AI Actually Touches Money
How Amazon uses LLMs to classify damage and set refunds — and the guardrail stack that keeps the AI from touching the dollar amount directly.
#4834: Why Your AI Bill Keeps Surprising You
How a 37-cent project and a $100 simulation reveal the hidden math of AI API pricing.
#4822: Cloud-Local AI Hybrid: Does It Actually Work?
Can you pair Claude's reasoning with a local Qwen 7B for savings? The gap is wider than you think.
#4792: Shadowing an IMS Forecaster in 2026
What does a human forecaster actually do when AI models already handle 95% of the routine work?
#4789: How LLMs Actually Know When to Stop
Three layers of stop signals, from EOS tokens to context limits — and why none of them really "know" when to stop.
#4788: How Weather Prediction Became a Science
From hand calculations to chaos theory to machine learning — the full arc of forecasting.
#4778: SSH MCP and Tailscale: Docker Migration Guide
How to configure Tailscale for Docker containers when migrating from public IPs to tailnet addresses.
#4774: When a Token Isn't a Word Anymore
Images, audio, and code all use "tokens" — but they're completely different things. Here's how tokenization actually works across modalities.
#4760: Context Window Management: Cutoff vs. Compaction
Trailing cutoff or auto-compaction? How to manage Claude Code's context window without losing critical information.
#4723: Are Model Families Actually Different Models?
Claude Opus, Sonnet, and Haiku aren't trim levels — they're different models sharing a brand name.
#4717: Building a Home Whisper Server: The Full Spec
Daniel wants a dedicated home server for Whisper dictation. Here's the exact build — GPU, RAM, storage, and runtime — to make it work.
#4696: Why AI Over-Explains Simple Tasks
Why does AI turn a simple request into a 24-page document? We explore the training biases and architectural limits behind AI's tendency to over-del...
#4690: Typst for Agents: Programmatic PDF Generation
How Typst's function-based design makes it the ideal document engine for AI agents generating professional PDFs.
#4686: How Claude Code's Patience Works: Fork, Wait, and the Transcript Stack
Claude Code's "waiting" is an illusion. We break down the Unix primitives behind its background task system and where the analogy cracks.
#4683: Parallel vs Sequential Agent Loops: The Real Tradeoffs
When should AI agents run in parallel vs sequential chains? The dependency graph isn't enough — join costs and error handling matter more.
#4669: What "Distribution" Really Means in AI Models
Unpacking what "distribution" actually means under the hood — and why training data variety shapes model behavior.
#4605: Why Chatterbox Still Leads Open-Source TTS
We break down the architecture of Chatterbox, the open-source TTS model powering this podcast, and why it remains state of the art.
#4604: When Sub-Agents Overwhelm the Orchestrator
Daniel's inventory system hit a wall when parallel sub-agents flooded the orchestrator's context window. We break down the math and the fix.
#4588: The Hidden Cost of Conversational AI: Why Stateless LLM Calls Matter
Why conversational AI tools are making batch document processing slower, costlier, and less reliable — and what to do about it.
#4581: Text In, Text Out: Designing Single-Purpose AI Utilities
Why small, narrowly scoped AI tools need system prompts that say "shut up and return the output" — and how to write them.
#4580: The Tiny AI Tools That Beat Big Agents
Why small, single-purpose AI utilities outperform autonomous agents for daily work — and what interface they need.
#4523: Can You Trust an AI's Summary?
Dedicated text compressors exist but aren't in production. The real problem? Nobody can verify the summary didn't drop what mattered.
#4492: What Makes a Container Robot-Ready?
The hidden specs that turn a plastic box into something a robot can pick, read, and move at industrial speed.