# Evident Machines

> Evident Machines benchmarks frontier AI models on real research and engineering workflows in electronics, electrical engineering, and mechanical engineering.

## Frontier models are opening new possibilities in electrical engineering.

Much of scientific discovery in electrical engineering happens through physical experiments, where real-world observations and measurements reveal information a simulation didn’t capture. This workflow is inherently iterative, moving between theory, computation, and the physical lab bench as the research develops.

We evaluate how models exercise scientific judgment throughout that process. Their ability to design a research protocol, investigate an unexpected result, and weigh competing constraints will shape how much research we can entrust to them.

## What we do

We build benchmarks around long-running research and engineering projects. The tasks bring together specialized software, experimental measurements from the lab bench, and physical constraints.

We hope to create a clearer picture of scientific capability in electrical engineering. By making failures specific and reproducible, we aim to provide concrete targets for model improvement and a way to measure whether those improvements carry through to research at the lab bench.

## Work with us

We’re looking for scientists, engineers, and researchers in electrical engineering to help build meaningful evaluations from work they know deeply. If you’re interested in shaping the benchmark or collaborating on a research task, get in touch.

To contribute to our benchmark paper, see our call for contributions.

## Call for contributions

We’re looking for researchers and engineers with hands-on experience in electronics and electrical engineering to contribute to AI research. You’ll help develop practical engineering problems drawn from research, lab work, or industry. The work includes hands-on experiments, documenting methods and results, and reviewing technical contributions with our team.

Researchers, graduate students, and practicing engineers are welcome. You should have experience designing, building, or testing hardware and be able to explain your methods and assess whether a result is technically sound.

Compensation will be determined by background, expertise, and geographic location. You’ll receive co-authorship on our benchmark paper if your task is accepted into the benchmark.

## Pages

- [Home](https://evidentmachines.com/): Evident Machines benchmarks frontier AI models on real research and engineering workflows in electronics, electrical engineering, and mechanical engineering.

- [Call for contributions](https://evidentmachines.com/call-for-contributions/): We’re looking for researchers and engineers with hands-on experience in electronics and electrical engineering to contribute to AI research. You’ll help develop practical engineering problems drawn from research, lab work, or industry. The work includes hands-on experiments, documenting methods and results, and reviewing technical contributions with our team.

- [Privacy Policy](https://evidentmachines.com/privacy/)

- [Terms of Service](https://evidentmachines.com/terms/)