The St. Louis Fed provides a free API for 800K+ economic data, and I built an MCP that plugs it into Claude, which just takes 5 min to set up: https://t.co/gww6FuNY0X
I asked it the recession odds in 2026. After pulling a dozen recession indicators, it came back with 20% 🙏
"Think from first principles" might be the highest-leverage prompt I know. Without it, AI just summarizes whatever it scraped. With it, the quality jumps an order of magnitude. Just asked Claude about the endgame of coding agents—the answer was excellent.
@lpolovets It doesn’t look like we need a lot more software, so that means the TAM only expands when more engineers are replaced. If AI replaces 50% of engineers, each making $200k now and AI captures 20% of value, that’s $400b.
I finally wrote another blogpost: https://t.co/WddJkbSfks
AI just keeps getting better over time, but NOW is a special moment that i call “the halftime”. Before it, training > eval. After it, eval > training. The reason: RL finally works.
Lmk ur feedback so I’ll polish it.
My NEW Official Trump Meme is HERE! It’s time to celebrate everything we stand for: WINNING! Join my very special Trump Community. GET YOUR $TRUMP NOW. Go to https://t.co/GX3ZxT5xyq — Have Fun!
What's expression language, and why do smart teams choose expression language to implement rules? Learn more in our latest blog post.
https://t.co/TwQ6oLR1GY
We're excited to announce integrations with leading fraud, credit, and compliance data providers!
In addition to saving time on integration, our customers orchestrate data sources through workflows to cut latency and costs.
https://t.co/qmSjz96DhW
I'm blown away by the versatility of @NotionHQ again. We just launched our public docs with Notion and @super_ (custom domain, SEO) in hours instead of weeks.
No-Code boosts productivity, and Sperta's customers agree too. Check out our results at https://t.co/G2fNTGZncW
Introducing experiments (a.k.a A/B testing), which complement backtesting and help you establish causation instead of correlation.
With these testing capabilities, Sperta further helps you optimize fraud, credit, and compliance decisions.
https://t.co/xMoIqBlRHc
To give you more confidence in the backtesting results, we designed it to be a distributed and scalable solution, which is especially important for imbalanced datasets in use cases such as fraud.
We're excited to launch backtesting! Backtesting lets you test a new workflow version against historical data to gain more confidence before putting it into production.
Read our blog to learn more:
https://t.co/wdMkXOn0LJ
We are excited to announce our 6.4M seed round led by @Lightspeedvp to bring modern observability to decentralized applications. @HashKey_Capital Canonical(@ai) Essence(@tnachen) @GSRVentures also participated in the round https://t.co/f73FHg5Rm7
Sperta is part of the core workflows of our customers, so we're serious about maintaining high availability and low latency.
We are now promising 99.9% availability, and we also released three new features to improve decision quality and safety!
Sperta entered GA with a 99.9% availability SLA. Now, you can also improve decision quality and safety with rule versioning, shadow testing, and percentage rollout.