A supercomputer at your fingertips
I'm now up to five internal Options Technology key note and town hall presentations on "AI transformation" in
AI for Financial Services.
Clients asked for a way to keep up. AI moves fast, the landscape shifts weekly, and it is hard to separate signal from noise. This is where we write down what we learn: architecture decisions, benchmark results, model releases, and field notes from building private AI on bare metal.
Latest
Introducing PrivateMind Jobs
Chatting to hedge fund clients over the last few weeks the feedback was that the only thing preventing them from moving all
What Machiavelli Can Teach Us About AI: The Effective Truth of the Cathedral
"Everyone sees what you appear to be, few experience what you actually are." – Niccolò Machiavelli, The Prince In 1513, Machiavelli
The Trust Wall: Why AI Robo Workers Will Run on Sovereign Models
The era of the AI Robo Worker may be on the horizon, but Fortune 500 companies are quickly figuring out that sovereign,
Powering the Future of Finance with High-Density Infrastructure
At Options, we see firsthand how rapidly the financial services landscape is changing. For years, the primary currency of electronic trading was
Open Models, Sealed Networks: Why Enterprise AI Value Is Moving Up the Stack
It's a good time to remind ourselves of Alex Karp and Palantir Technologies' experience with locked-down open models. It
The Frontier Model Honey Trap
The drama around Navier-Stokes is breaking the internet. If you've been following my posts, you'll know that my
Your Enterprise AI Doesn’t Know Your Company. It Just Searches It.
RAG gave companies a better search box. The industry started calling it memory. Now agents are exposing the difference — and the real
BuffettBot got an upgrade
We rebuilt BuffettBot end to end: a full retrain that thinks in Buffett's own principles, retrieval over 55 Berkshire documents, and five financial-data tools it never saw in training. The old model had no working tool interface; the new one completes all twelve scenarios.
What daily bars hide: why our market data is the edge
We ran CVaR on 2,625 real one-minute bars from the Atlas Feed. Daily bars make intraday flash crashes invisible. The data is the edge. The optimiser just needs to be fed.
Using LLMs for blogs: Cheating?
We ran the same prompt through four model aliases. One scored 33% AI on GPTZero. The others scored 100%. Here is what gives it away.
OpenAI's 80% price cut, and what it doesn't change
Three weeks after GPT-5.6 launched, OpenAI cut its two cheaper tiers by up to 80%. The official story is efficiency; the list is really converging on open weights. The price was never the moat.
Stay current
Clients asked us for a way to keep up with AI without the noise. This is it. Posts on sovereign AI, GPU infrastructure, open-weight models, and what the news actually means — by email, or in your feed reader.