We Rebuilt Alphanume From Scratch

We’re not going to become the best data vendor in the world by standing still.

For the past few months, one of the most common pieces of feedback we received about Alphanume sounded something like this:

“The datasets are incredibly interesting, but APIs are too difficult to use.”

Which, honestly, was fair.

The original version of Alphanume was built for people already living inside quant infrastructure stacks: Python environments, cloud jobs, point-in-time backtests.

For long-time systematic traders and developers, that worked fine. However, it created a quiet bottleneck.

Plenty of people understood the market mechanics perfectly well and still couldn’t explore the datasets without engineering around them first.

That kind of operational friction is normal inside institutional quant workflows, but the problem was that it had become the thing standing between the research and the trader.

So, we rebuilt the entire platform from the ground up.

Not just the homepage, not just the docs, but the entire platform.

Every dataset on Alphanume now has a fully interactive dashboard with full access directly in the browser. You can filter, inspect, search, visualize, and explore the datasets immediately without writing a single line of code:

The API still exists, naturally, and nothing about it changed (if you have already built around it, you lost nothing). It’s just not the only door in anymore.

Same datasets, same point-in-time histories, same alpha surface underneath. You just get to it directly now.

If you want to inspect historical SPX 0-DTE strike bands against realized intraday movement, that is a browser tab. Dilutive SEC filings, lifecycle resolution, market cap at filing, event persistence, same thing. Filing-intensity spikes, de-SPAC completions, corporate-default timelines, point-in-time momentum constituents, the volatility regime classifier, the optionable-universe snapshots, and the list goes on.

Now, these are not gimmick datasets dressed up with nicer charts. They’re the same datasets sitting underneath a large part of the research we publish: the strike infrastructure, the dilution and corporate-distress systems, the prediction-market work, the cross-sectional momentum baskets, all of it.

Most of this data does not exist anywhere else in clean, point-in-time form, and that’s why we started this in the first place.

To everyone who has been here through the API-only era, querying endpoints and building around the rough edges: thank you. You were using this platform when using it took real effort, and a lot of what got rebuilt got rebuilt because of what you told us.

We are not slowing down on the part that actually matters. The proprietary data front is where the work continues, and the quantitative trading experiments we run here every week feed that engine directly. Every strategy we test surfaces a question, and the good questions become datasets. That loop is what positions us to keep building some of the most predictive data out there.

And to everyone who opens these posts when they land and gives us your time: we do not take it for granted. Thank you for coming along.