i think gambling could actually be one of the best fits for Jev
most bets are already bounded decisions:
team A or B
over or under
yes or no
Jev doesn’t need to generate anything, it can compare the options, score them, and return probabilities extremely fast.
add live stats, odds, injuries, history and news, and you get a much better decision layer
not a guaranteed edge, but a very natural fit for how Jev works
building your own Jev is stupid!
you can copy the scoring trick, sure
but then comes training, calibration, evals, infra, tuning, maintenance…
if you’re doing research, go for it.
if you’re trying to ship a product, rebuilding all of that is just wasting time!
especially when using Jev costs almost nothing.
with Jev you can build and test pages much faster using the components you already have
perfect for prototypes, internal tools and fast product experiments
@MrBrjan true,
but i see it more as a way to move faster with your own system, not replace design decisions completely
for example
great for quickly building an admin page or testing a new onboarding flow with components you already trust.
@DopeRakshith fair distinction.
in this demo both models are selecting from the same component system, so the point is the decision + render speed, not generating raw HTML from scratch
we had a big research for tracking some realworld object with AI the problem was speed
jev might fix that need to rewrite some things i tested few months ago going to share
🧵 I gave Jev $10,000 and let it trade BTC again.
But this time, I gave it everything a trader would look at:
market data, derivatives, macro, on-chain data, news and sentiment.
30 days
🧵 I gave Jev $10,000 and let it trade BTC again.
But this time, I gave it everything a trader would look at:
market data, derivatives, macro, on-chain data, news and sentiment.
30 days
That’s the part I find most interesting
Jev isn’t just reading headlines
It’s trying to understand how an event flows through the market like an read trader!