Our journey
We built it for ourselves first.
Instructed Intelligence began as a portfolio of ventures run almost entirely with AI. We hit the fragmentation problem before we had a name for it — every session sharp, nothing accumulating — so we built the knowledge layer that fixed it. Then we found we were not the only ones who needed it.
Run on the platform we sell
A small, agile team delivering scale and quality far beyond its size.
1,146
Pieces of work tracked
Every ticket, story and decision coordinated through the platform rather than through anyone's memory.
28
Projects
Products, client engagements and internal work, all running on one shared knowledge base.
360k+
Lines of code committed
28 repositories. Equalling 164 developer years.
Counted on 13 August 2026 from our own Cornerstone AI Coordinator.
The belief
Most enterprise software exists to work around human limits.
Filing, finding, summarising, proving. A large part of every estate is there because nobody can hold what the organisation knows in their head. For decades that was the right answer to a real constraint.
AI does not lift that constraint while it remains a guest in those systems. It reads whatever it can reach, writes nothing back, and starts again tomorrow.
Which is why individuals get faster and the organisation does not.
on a fresh task, with an AI pair-programmer.
Peng et al., 2023
on real work in code they had known for years — while believing they were 20% faster.
METR, 2025
delivery, for every 25% rise in AI adoption, while teams reported feeling more productive.
DORA, 2024
Everyone got faster. The organisation did not.
Our vision
The estate gets smaller, not larger.
An AI-native organisation is a by-product rather than a programme. The more the organisation works, the more it understands itself, until the tools that existed to help people file and find are no longer where the work happens.
What remains is what genuinely holds value: the ledger, the transaction, the system of record. What falls away is the layer built to compensate for memory.
One layer, in every industry where getting it wrong is expensive — each enterprise owning what is in its own.
Our mission
Build the layer the rest of the estate can run on.
One shared layer, fed by every surface: the knowledge a business owns, typed, cited and versioned — and owned by the enterprise, not rented from us.
Start where the pain is sharpest. Let it compound. Let the rest become optional.
The team
Leadership
Jason Gould
Technology & product
Designed and built Cornerstone end to end.
- 30+ years shipping software at scale in regulated industries: trading and core banking for two of the world's largest banks.
- Led the cloud-native rebuild of one of the world's largest retail estates; drove a 49% first-year rise in online sales as its Chief Technical Evangelist.
- Forbes contributor on software and organisational transformation.
James Washer
Product & commercial
Owns product, methodology and commercial framing.
- Serial founder: 16+ years starting, scaling and exiting B2B SaaS and e-commerce businesses.
- Built photo-commerce platforms partnered with global consumer and print brands, embedded in apps with 15m+ daily users.
- Operating Partner at a tech-focused family office, leading product and M&A across its portfolio.
Edward Durell
Commercial & finance
Owns commercial, financial and corporate discipline.
- A decade at a Big Four firm in assurance and audit.
- Group CFO/COO and MD roles across consumer goods, sport, food and financial services.
- Two decades making businesses ready for the scrutiny of an enterprise customer or an investor.
Careers
Work for Instructed Intelligence
Everything on this site was built and shipped by a team working this way. That is not a boast about hours — it is what the platform is for, and it is the condition of joining us.
We hire rarely and deliberately. There is no graduate scheme, no bench, and no layer between you and the people who decide. If you join, you own a lane.
What we look for
People who finish things
The hard part is not starting. We value the last ten per cent — the tests, the migration, the honest note about what did not work.
Accountable for what AI produces
Everyone here works alongside AI daily and is accountable for what it produces. Being fast is not the point; being right and being able to show why is.
Writing things down
Our knowledge is a shared asset, not a private one. If you would rather explain it once, well, than five times badly, you will recognise how we work.
No formal openings are posted. If the above sounds like you, write to us anyway and tell us what you would own.