Location
Menlo Park, CA, United States
Posted
July 11, 2026
Job Description
Description
As a ML Engineer, you'll be joining this client as the begin building the next generation of Computer Use Agents (CUA). You'll own benchmark development and model failure analysis for the CUA training pipeline. You'll be running evaluations against model checkpoints, analyzing where and why the model fails, and feeding those insights back into the training loop.
Responsibilities:
* Land new benchmarks and eval suites for the CUA models
* Run model evaluations against checkpoints and analyze results
* Perform model failure analysis — dig into where the model breaks down and why
* Write Python and SQL to process, transform, and analyze large-scale data
* Support automation of eval workflows over time
* Collaborate with research scientists, engineers, and TPMs on the team
* Document methodology and results
* Communicate findings clearly to stakeholders
Top Skills
python, SQL, Data Analysis, Machine Learning, Model Evaluation, Benchmarking, F...
As a ML Engineer, you'll be joining this client as the begin building the next generation of Computer Use Agents (CUA). You'll own benchmark development and model failure analysis for the CUA training pipeline. You'll be running evaluations against model checkpoints, analyzing where and why the model fails, and feeding those insights back into the training loop.
Responsibilities:
* Land new benchmarks and eval suites for the CUA models
* Run model evaluations against checkpoints and analyze results
* Perform model failure analysis — dig into where the model breaks down and why
* Write Python and SQL to process, transform, and analyze large-scale data
* Support automation of eval workflows over time
* Collaborate with research scientists, engineers, and TPMs on the team
* Document methodology and results
* Communicate findings clearly to stakeholders
Top Skills
python, SQL, Data Analysis, Machine Learning, Model Evaluation, Benchmarking, F...