Sayre Blake
A demo can do it. Can the product depend on it?
A successful model demonstration shows a capability. A product has to deliver its intended behavior across many different encounters.
Updates
A selective chronological record of writing, appearances, project-linked notes, and art from across the work.
Sayre Blake
A successful model demonstration shows a capability. A product has to deliver its intended behavior across many different encounters.
Sayre Blake
Five principles for using language models as processors of meaning while keeping reliable state, rules, and verification in the surrounding system.
AI Practitioners Summit
I was named AI Rising Star of the Year in the 2026 AI Practitioner Awards.
Sayre Blake
A model can propose what someone meant. The system has to decide whether that proposal becomes a lasting change.
Sayre Blake
If the model is one bounded processor inside a larger system, the model can remain probabilistic while the system becomes testable again.
AI Practitioners Summit
My AI Practitioners Summit talk on designing systems around the work language models do well, with reliability, cost, and human judgment in view.
OnlyFounders
Anne Gillaspie and I open a recurring series on the culture around AI, examining how human decisions, corporate narratives, and system design shape the technology we use.
Sayre Blake
A language transform earns its place where a designed system encounters meaning that its ordinary rules cannot use directly.
Sayre Blake
Evaluate whether a language transform preserves the distinctions the system needs, then test the system's rules separately.
Sayre Blake
A language transform can receive the context it needs for one operation while the system remains responsible for memory and state.
Channel Insights AI
Dinara Bakirova and I discuss why soft skills grow through repeated practice, how SkillSaige uses short sessions, and what a founder has to decide when building with AI.
Sayre Blake
Design the behavior people need first. Then ask whether a stateless language transform makes one part of that system better.
Sayre Blake
If you've even considered using AI for business (or really production software of any kind), you should probably be asking this question.
Jade Ember Studios
FleXss brings stylesheets, design tokens, theming, and modern layout workflows to Unity through one renderer-independent UI framework.
Chasing the Insights
A conversation with Vince Warnock about looking beyond the chatbot as we design useful AI systems and practical learning experiences.
Sayre Blake
A practical look at why LLMs hallucinate, and what you can do about it (spoilers, "more prompts, more training" isn't the answer).
Break It Down Show
A return to Break It Down Show with Pete A. Turner and Brad Hutchings to examine language models as coprocessors and the case for more humane, sustainable AI systems.
Her Success Story
A conversation with Ivy Slater about building where art meets engineering, using AI with discipline, keeping teams lean, and choosing investors with intention.
Mike Schiano In the Queue
A practical conversation with Mike Schiano about bounded AI, repeatable workplace practice, and why chatbots and passive training rarely create behavior change on their own.
Sayre Blake
“We’ve made a computerized screaming void of the collective human subconscious. Would you like to talk to it?” What LLMs become when you take a step back, and why it's actually not that scary in the light.