Machine Learning Research Engineer

Responsibilities

You will be responsible for the evaluation layer of our expert intelligence factory. While our pipeline handles the raw telemetry capture, you will build the benchmarks and scoring systems that determine if an agent actually learned the "dark matter" of professional workflows. This means designing metrics that go beyond simple pass/fail to measure the fidelity of an agent's motor control and its ability to handle real-world UI friction.

We are looking for someone who understands that the biggest bottleneck in agentic AI right now isn't just getting more data, but proving that the data actually works in production.

• You must be a current PhD student in Computer Science or a relevant field.
• You need prior industry or research experience building agentic evaluation frameworks, benchmarks, or scoring infrastructure, ideally from a place like Surge or a frontier AI lab.
• You must be a US Citizen to meet our current project and compliance requirements as we are unable to accept F-1 or other visa statuses for this role.
• You can formally commit 20 to 25 hours per week for this remote or hybrid position at a pay rate of 40 to 50 dollars per hour.

Entry level

Part-time

Engineering and Information Technology

Software Development

Referrals increase your chances of interviewing at Archal Labs by 2x.