Research where it can become engineering.
Skyneski explores technology at the point a question can be tested and built. The output is a system, not a claim.
R&D approach
Each step exists to stop a promising idea from skipping the evidence. Deployment is where the next question usually appears.
- 01Research
A question narrow enough to test
- 02Experiment
Change one variable and watch
- 03Prototype
The smallest build that can be wrong
- 04Validate
Does it hold outside the demo
- 05Engineer
Make it observable and operable
- 06Deploy
Learn from a system in use
Research areas
These are directions of work. They are not a list of completed studies, publications, or funded programmes.
- 01
Connected intelligence
Systems that join a physical signal to a later decision without losing the meaning of the signal.
- 02
Predictive systems
Estimates of what may happen next, kept as estimates.
- 03
Cyber-physical systems
Software that has to respect a device, a place, and a failure in the real world.
- 04
Infrastructure intelligence
Using operational data to understand compute, network, and reliability.
- 05
Data-driven decision systems
Analytics that prepare a judgement and leave the judgement with a person.
- 06
Applied AI
Models placed inside a system that can be operated and questioned.
- 07
Industry-specific technology
Platforms such as Horse Monk, Race Monk, and Horse Commerce, built from the same practice.