Artificial Intelligence
Episodes whose main subject is AI itself: building with, training, evaluating, or debating AI systems. AI used only as a tool for another topic goes to that topic.
#2511: Measuring AI API Latency Through the Black Box
How to benchmark token throughput and debug slowdowns in closed CLI tools like Claude Code using OpenTelemetry and mitmproxy.
#2510: The Design That Makes Voice Agents Tolerable
Drive-thru accuracy, healthcare triage, and the design secret that makes people *want* to talk to a machine.
#2507: The AI Design Engineer: Your New Job Title?
What happens when product thinking meets AI agents? The future of software work is here.
#2505: How Self-Hosted Search Actually Works for AI Agents
SearXNG isn't a crawler — it's a metasearch router. Here's how it works and why AI agents change everything.
#2499: Coding by Voice: Teaching a Beginner TypeScript
Learn to code by building a real TypeScript tip calculator — no experience needed, just your ears and keyboard.
#2496: Are Hidden API Endpoints Leaks or Just Plumbing?
When LLM agents discover unauthenticated JSON endpoints in browser DevTools, is it a security breach or just reading the page?
#2495: How to Bake Personality Into an LLM in 15 Minutes
Fine-tune a model's personality with ~300 examples and a consumer GPU. SFT + DPO explained.
#2494: Active Prompt Engineering: Daniel's Diff-Based Loop
A deep dive into iterative prompt refinement using inter-iteration prediction change as an uncertainty signal.
#2493: Are You Writing for Humans or AI Agents?
How GitHub repos, JSON formats, and competing standards are reshaping who (and what) you're publishing for.
#2492: When AI Agents Collapse Stack Evaluation from Weeks to Seconds
How Claude Code and agentic AI are turning GitHub into a discovery layer and collapsing library evaluation from weeks to seconds.
#2488: Hybrid Pipelines for Entity Resolution
Classic NLP pipelines vs. lightweight LLMs for handling Hezbollah’s half-dozen spellings.
#2487: Why AI Calls Everything a "Prediction" (Even Images)
Machine learning calls everything a "prediction" — even generated images. Here's why the terminology matters more than you think.
#2486: Why Noise Reduction Can Ruin Transcription Accuracy
Cleaning audio before transcription can increase errors by up to 46%. Here's the right approach for your voice app.
#2483: Substitution Anonymization: Privacy Without Utility Loss
How to generate realistic synthetic voice notes and calendar data with zero PII exposure risk.
#2482: When AI Chatbots Leak Your PDFs via Public S3 Buckets
A user uploaded a sensitive PDF to an AI chatbot. The chatbot stored it in a public S3 bucket with zero authentication.
#2479: The Screaming Baby Stress Test
Choosing the right headset and control method for dictation when you're holding a baby who won't stop screaming.
#2478: MCP File Handling: Why Your Base64 Upload Breaks at 4MB
MCP has no standard file input. Base64 breaks at 4MB, presigned URLs need whitelisting, and MinIO workarounds aren't standardized.
#2472: When Guardrails Break: The Hidden Costs of AI Gateway Filtering
PII detection at the gateway layer can block legitimate invoices. Here's how guardrails actually work and where they fail.
#2471: Creative Briefs for AI Agents: What Agencies Already Know
How agency best practices for briefing creatives map directly onto getting reliable output from AI agents like Claude Design.
#2470: Where Intelligence Should Live in Your Pipeline
When should you fine-tune a tiny model for prompt enhancement instead of prompting a large one? The answer depends on latency, precision, and domain.