All Episodes
Page 12 of 320 · October 2026
#6303: Blind Testing AI Video Models for Architects
A matrix of sixteen prompts and two seeds across three video models produces ninety-six clips to generate, fund, and review.
#6302: Why Fine-Tuning Whisper Won't Stop Duplicated Text
Whisper evaluates transcripts against a gzip compression threshold of 2.4, rejecting repetitive text before retrying at higher temperatures.
#6301: Building an AI Rubric to Rank Israeli Parties
A customized AI rubric ranked Israeli political parties by prioritizing tenant rights over the headline national issues dominating the campaign.
#6300: Choosing the Right Audio Annotation Tool
Desktop tool ELAN supports independent timeline tiers to prevent overlapping sounds like car horns from flattening into speech transcripts.
#6299: Finding Recurring Urban Sounds With Four Examples
In DCASE evaluations, prototypical network baselines can locate specific sound events across long recordings using as few as five labeled samples.
#6298: Why Dictation Models Steal Other People's Voices
Stock speech models lack target-speaker conditioning, defaulting to transcribing whichever competing voice is loudest or most predictable.
#6297: Teaching Dictation Models to Ignore Other Voices
Evaluating Whisper on non-speech clips revealed entire invented phrases in roughly one percent of transcriptions.
#6296: Why Speech Recognizers Cut Off Hesitant Speakers
Between speakers, conversational handover averages 200 milliseconds, but planning pauses inside a turn naturally take seconds.
#6295: When Splitting Serverless Apps Earns Its Keep
Modal apps enforce atomic deployment, meaning all grouped functions redeploy in place together while scaling independently.
#6294: Why Tool Calls Break Across Coding Models
Translating coding agents across providers fails when input schemas, user-role results, and tokenizer markers diverge.
#6293: Ann McCann on What Dental Hygiene Teaches Us About Benchmarks
A 1990 experiment showed switching from letter grades to pass-fail changed student motivation without altering clinical errors.
#6292: Mark Mehlman on Predicting Random Signals When Statistics Drift
A harmonizable process is mathematically identical to an orthogonal projection of a stationary process living in a larger space.
#6291: Surplus Vote Math in the Knesset
Under the Bader-Ofer method, party surplus agreements distribute marginal Knesset seats without transferring votes across rival candidate pools.
#6290: Polymarket's Final Odds for Israel's 2026 Vote
Prediction markets put Eisenkot at 55.9 percent for prime minister while giving an outright majority only a 16.5 percent chance.
#6289: Why Hermes and Pi Top the OpenRouter Coding Table
Hermes Agent and pi logged 1.66 trillion and 1.41 trillion tokens on OpenRouter, topping Claude Code despite fundamentally opposite designs.
#6288: What Happens Between Model Checkpoints
Extending reinforcement learning on Olmo 3 Think for 21 days yielded a 20-point leap on IFBench without starting fresh pretraining.
#6287: Why Model Names Conceal Changes Behind the API
One API identifier spanned ten distinct releases over nineteen months without changing the name submitted by the customer.
#6286: Why Claude Code Replaced Vector Search with Grep
Claude Code creator Boris Cherny revealed early prototypes used local vector search before abandoning it for agentic grep.
#6285: Why Coding Benchmarks Split Across Models
GLM 5.3 claims 88.2 on Terminal Bench 2.1 against Opus 4.8 at 85, but newer evaluations reveal a very different order.
#6284: Why Coding Agent Configurations Do Not Travel Well
Codex caps combined instruction files at 32 kibibytes, which can crowd out nested rules before an agent even starts work.