Nathaniel Whittemore

speaker
51,694 appearances 202 recordings 3 series first heard Oct 2025 last heard 3d ago

Nathaniel Whittemore’s voice in public audio — every appearance, attributed to the second.

Trend

recordings per month · last 12 months
31 · Jun OctJan 26AprJulnow

Recordings per month over the last 12 months — 202 in all, peaking in Jun 2026 with 31.

Appearances

newest first · ▶ plays the moment
As Chubby sums up, JEV is an AI model built for decisions rather than text generation. podcast-host
And it is important to note that the trade-off here is that this model does not generate text. podcast-host
It is not, in other words, a replacement for LLMs in general. podcast-host
It's a replacement for a certain category of work that LLMs do, where they've been very square peg mashed into a round hole to do it. podcast-host
The idea continues Chubb. podcast-host
is to embed fast, cheap AI decisions into software. podcast-host
So what is this model actually built for? podcast-host
Well, think of how much of office work consists of reading something and deciding what should happen next. podcast-host
Does this message need a response? podcast-host
Which department should handle it? podcast-host
Does this document answer the question? podcast-host
Is this customer describing a bug or asking for a feature? podcast-host
Does this draft make a claim its source doesn't support? podcast-host
Is the situation routine enough to automate or should someone review it? podcast-host
These are judgments about meaning, and they're often difficult to express as fixed rules. podcast-host
JEV then is designed to take the relevant information and answer narrowly defined questions with probabilities, categories, or scores. podcast-host
The surrounding software then uses those answers to route, rank, flag, or proceed. podcast-host
In the TypeSafe documentation, they explicitly recommend breaking complex decisions into small questions and then combining their results in code. podcast-host
So where would this show up in normal business? podcast-host
One obvious area is customer support, where the small judgment the model could make would be something like, is the customer frustrated? podcast-host
Showing 2681–2700 of 51,694 · page 135 of 2585 ← Previous Next →