Comments - Clouded Judgement 3.20.26 - Digital Twins
Clouded Judgement
SubscribeSign in
Clouded Judgement 3.20.26 - Digital Twins
Every week I’ll provide updates on the latest trends in cloud software companies. Read →
3 Comments
Top first
Digital twins are the missing layer -- agreed completely. But the infrastructure gap beneath them is even bigger than most realize. A digital twin of a factory or fleet is only as good as the real-time data feeding it, and that means solving the IoT connectivity problem: thousands of sensors, robots, and edge inference nodes need deterministic low-latency networking, eSIM-based carrier orchestration for mobile assets, and OTA firmware governance to keep the physical-digital sync accurate. The companies building the connectivity and edge data pipeline layer between physical assets and their digital twins will capture enormous value as this becomes the default enterprise AI architecture.
Reply
We have another use for this term. We use Digital Twins to test and validate software, especially code written by agents.
In previous regimes, a team might rely on integration tests, regression tests, UI automation to answer "is it working?"
We noticed two limitations of previously reliable techniques:
Tests are too rigid - we were coding with agents, but we're also building with LLMs and agent loops as design primitives; evaluating success often required LLM-as-judge
Tests can be reward hacked - we needed validation that was less vulnerable to the model cheating
The Digital Twin Universe is our answer: behavioral clones of the third-party services our software depends on. We built twins of Okta, Jira, Slack, Google Docs, Google Drive, and Google Sheets, replicating their APIs, edge cases, and observable behaviors.
More details at https://factory.strongdm.ai/
Reply
Have a look at Selflet.ai to see a digital twin factory
Reply