At AI Engineer Code Summit, Diogo Almeida Calls Jev a “Fourth Logic Gate” for Software
TypeSafe’s CEO rallied developers to embed System One models directly in application code, followed onstage by Glean and UiPath showing production routing and sub-second ranking.
System One models took center stage at the AI Engineer Code Summit on November 11, where TypeSafe AI co-founder and CEO Diogo Almeida delivered a keynote urging developers to move beyond fragile prompt chaining and embed calibrated judgment directly inside ordinary application code.
“You get Einstein 95% of the time, but Mr. Bean 5% of the time. And if you can’t tell which is which, how are you going to automate anything?”
Diogo Almeida, keynote at AI Engineer Code Summit (Nov 11, 2026)
Glean and UiPath show production patterns onstage
Glean: Zero-shot routing
Chinmay Goyal showed how Glean pairs Jev with frontier LLMs and classical classifiers so shifting customer taxonomies no longer require constant retraining—while keeping the model split completely invisible to end users.
UiPath: Sub-second tournaments
Chibi Vikram and Kevin Bernstein demonstrated a parallel document-chunk ranking “tournament” where thousands of Jev comparisons narrowed retrieved passages in under one second.
Almeida closed his keynote by pitching TypeSafe’s long-term goal: making System One judgment “as reliable as a fourth logic gate” inside everyday software.