Development
Building software: programming, developer tools, APIs, deployment and DevOps, and building with AI (agents, MCP servers, RAG, AI coding tools). An extra channel alongside an episode's main one, so AI engineering episodes sit in both.
#5991: Batch Inference on Your Own GPU: vLLM, SGLang, Ollama, llama.cpp
Hosted batch APIs sell you half-price tokens. Here's how to get the same file-in, file-out workflow on your own Linux box.
#5982: Teaching a Voice Engine to Say "Yalla" Right
A small token classifier that spots Hebrew words hiding in English sentences — and why finding the word is the easy half.
#5981: Batch Inference: Four Doors, Half Price
Same tokens, half the price — and almost nobody using it properly. A tour of batch inference from OpenAI to OpenRouter.
#5975: Merging vs Distillation: Can Two Models Become One?
Merging and distillation get lumped together, but they're opposites. One composes existing capability; the other has to teach it.
#5973: MCP Security: Tool Poisoning and Agent Identity
MCP servers trust their own tool descriptions. That design choice creates a whole new class of attack — and a new observability problem.
#5970: Swapping Your Linux Kernel: What Actually Changes
Your kernel is swappable without reinstalling anything. But do Liquorix and XanMod actually beat the default? The benchmarks say less than the mark...
#5965: Compiling From Source: What You Actually Get
Source builds aren't faster — they're smaller. What USE flags, compiler flags, and the build feedback loop actually buy you.
#5964: Containers vs VMs for GPU Inference: What You Actually Pay
Containers cost 0.03%. Passthrough VMs cost 1%... or 36%. Same hardware, different systems — here's how to tell which you'll get.
#5963: Running One Model Across Mismatched GPUs
Tensor, pipeline, or offload? What actually happens when you run one model across several GPUs — including mismatched ones.
#5961: What Your GPU Is Actually Waiting On
Your GPU says 100% utilized. It's multiplying zeros. Here's what the counters actually measure during inference.
#5960: There's No Such Thing as Tokens Per Second
Why one tokens-per-second number can't describe an LLM server — and which metrics actually tell you if it's working.
#5959: What RTF Actually Tells You About Speech Models
Accuracy gets the headlines, but speed decides what ships. How RTF, latency percentiles, and memory shape real ASR deployments.
#5957: The Scientific Plugins Hiding in Plain Sight
ChatGPT's plugin directory has 5,723 entries — but the real scientific AI work is happening in arXiv papers, not plugins.
#5956: Will ChatGPT Eat Your Grocery App?
MCP servers, ChatGPT apps, and a prediction that AI vendors will absorb your grocery list one micro-app at a time.
#5950: Self-Hosting Your Search Stack: SearXNG, Crawl4AI, and the Real Bill
Can SearXNG, Crawl4AI, and an open-source aggregator really replace a commercial search API? Yes — with a very large asterisk.
#5949: Inside Deep Research Agents: Claims, Citations, and the 84.7% Orchestrator Problem
Why 84.7% of citation errors originate at the orchestrator — and how a claim ledger fixes it.
#5948: LangGraph, LangChain, and Deep Agents in 2026
Runtime, framework, harness — how three LangChain layers actually fit together, and what changed in 2026.
#5945: How a Reward Model Actually Trains an LLM
One number, no gradients, four models in the room. What a reward model actually does inside the RLHF loop.
#5944: Keeping the Rollout Engine Fed: Inside RLHF Pipelines
Most of an RLHF run is spent waiting on generation, not gradients. Here's how TRL, OpenRLHF, and verl actually differ.
#5942: Service Workers: The Proxy That Never Sleeps
A JavaScript file with no tab, no DOM, and the power to intercept every request your site makes — and why refreshing never updates it.