Small team, unusually hard problems
Financial models are the messiest structured data in the world. Making them queryable without asking anyone to change how they work is the whole job.
| Ticker | Metric | Internal | Consensus | vs consensus | Last rev. | Next print |
|---|---|---|---|---|---|---|
| TSLATesla | Q4 EPS | $0.71 | $0.68 | +4.4% | 2d ago | Oct 22 |
| NVDANVIDIA | FY26 Revenue | $214.3B | $208.1B | +3.0% | 5d ago | Nov 19 |
| AAPLApple | Q1 EPS | $2.41 | $2.45 | −1.6% | 1d ago | Oct 30 |
| MSFTMicrosoft | Q2 Revenue | $69.9B | $68.7B | +1.7% | 3d ago | Oct 29 |
| AMZNAmazon | Q4 EPS | $1.62 | $1.58 | +2.5% | 7d ago | Oct 30 |
| METAMeta Platforms | FY26 EPS | $27.40 | $28.10 | −2.5% | 4d ago | Oct 28 |
Consensus, actuals and revisions out of the box
- MSFT
- AMZN
- GOOGL
- META
- ORCL
- NVDA
- AAPL
- TSLA
How we work
Depth over breadth
We would rather three things worked properly than ten things demoed well. Most of the product is invisible correctness: tracking that never loses an estimate.
Close to the user
Everyone talks to analysts. The gap between what a desk actually does and what software assumes it does is where this product lives.
Written down
Decisions, trade-offs and the reasons behind them get written down. A small team moving fast only stays coherent if the reasoning survives the week.
Where we tend to hire
No formal openings listed right now. Good people are hired when they appear.
Product engineering
TypeScript, React and a large amount of careful data modelling. Comfort with ambiguity matters more than any particular stack.
Data and research
Consensus data, revision history and the unglamorous work of making numbers from different sources mean the same thing.
Go to market
Talking to investment teams about their workflow without pretending to be one of them. Curiosity beats a pitch.
Nothing here fits but you think you should be here?
Send a note saying what you would want to work on. We read all of them.