Steps Ventures · writing
Blog
Short, plain-language pieces on what we are working on. Some are about our open research, the kind where we compute a signal out of public data and report the nulls too. Some are about the tools we build across energy, health, and real estate. We try to say what we know, what we do not, and how we checked.
Method · model choice · August 2026
The best model for reading a page depends on the script
On clean typeset print one generation wins by three times. On the nastaliq these lithographs actually use, the ranking flips and the earlier winner becomes the worst reader tested.
Read →Method · negative results · August 2026
The intervention that failed usefully
Cropping pages before reading them cost eleven times as much for no measurable accuracy gain. It shipped anyway, because it rescued eight percent of pages that returned nothing at all.
Read →Archives · ground truth · August 2026
There is no ground truth for Afghan lithographs
Nobody has hand-transcribed a page of this collection, so there is nothing to score against. Here is how you manufacture a benchmark, and what it cannot tell you.
Read →Method · benchmarks · August 2026
The forty pages that lied
Our benchmark file was sorted by difficulty. Taking the first forty rows was not a sample, it was a best case, and it inflated every absolute number we published.
Read →Method · cost · August 2026
Ninety percent of the bill was thinking that did not help
Cutting the reasoning budget saved ninefold with no measurable quality loss. Substituting a cheaper model looked identical on medians and lost 73 pages out of 94.
Read →Studio · how we work · July 2026
I do not write code. Here is what I shipped anyway.
A year of running a real software operation without writing code, through Claude. What worked, the honest failure rate, and the one habit that made it hold up.
Read →Ocean · eels · July 2026
How do you test a claim about a place no one has ever seen?
No human has watched a European eel spawn. Here is how you can still ask, with data, whether climate is moving its birthplace.
Read →Ocean · turtles · July 2026
Why we can trace a baby eel home but never a baby turtle
The ocean erases where a lost-years turtle came from in about two weeks. A baby eel is still traceable. One inequality says which animals fall on which side.
Read →Quantum · myth vs reality · July 2026
When does a quantum computer actually help with quantum biology?
The honest version. For the animal magnetic compass, a laptop still wins, today’s hardware is deaf to the signal, and we can name roughly the year that changes.
Read →Data · reproducibility · July 2026
A whole subfield’s accuracy numbers are too good to be true
Many published clinical machine-learning models report near-perfect accuracy that comes from a data mistake, not real skill. One correct change can drop a model from 0.98 to 0.79.
Read →Energy · AtomH2 · July 2026
What it takes to retire the diesel generator at a cell tower
Tens of thousands of European telecom sites still lean on diesel for backup. Solar plus solid-state hydrogen is a way out. A field study of where it pencils.
Read →Health · Selene · July 2026
Most supplements are built for an average that does not exist
A woman’s physiology changes across a 28-day cycle. A daily dose that ignores that is guessing. What it means to build for the cycle instead of the average.
Read →Energy · nLine · July 2026
There is energy leaving New York’s buildings as steam
Big buildings vent heat and pressure they already paid for. A public benchmarking record shows where recovery pencils, and where the filings do not add up.
Read →Studio · thesis · July 2026
The free thing has to stand on its own
How a one-person studio decides what to build. Give real value for free first, make the paid step obvious, and stay honest about what is a business and what is a project.
Read →
Subscribe: RSS · Atom · JSON. Every article is also served as plain markdown at /blog/<slug>.md, and all of them together at /llms-full.txt.