Episodes featuring Mike Taylor
Browse episodes where Mike Taylor appears as host or guest. View all episodes →
Mike Taylor, head of evals at Every, runs a live Think Week workshop on getting AI to write in your own voice. His method uses an automated test that checks which draft sounds more like your published work.
Mike Taylor, head of evals at Every, walks through DSPy with Drew and Isaac of Compound, the team behind it. They argue you should define tasks and examples, then let optimizers write the prompts instead of...
Dan Shipper and Natalia Quintero tour Every's frontier team, with Mike Taylor, Katie Parrott, Yannick, Kieran Klaassen and others showing how a dedicated group can test new AI models without distracting the rest of...
Every's team tries Google's new Gemini 4 Argon model and finds it surprisingly strong at writing and reasoning, though Google's fragmented product experience holds it back.
Every's Dan Shipper, Mike Taylor, Kieran Klaassen, and Natash live-test three new AI models across writing, coding, and design to help builders decide which to standardize on.
Every's team live-tests GPT-5.5 against Opus 4.7, covering where each model wins on speed, execution, planning, and design for builders and knowledge workers.