Senior AI Systems Engineer — Autonomous Agents (Palo Alto)

OpenKrill; United States; Remote; USD 200,000 - 300,000; Full time

Job summary

OpenKrill in Palo Alto, CA is building autonomous AI capabilities for executive productivity. You will develop agent capabilities and evaluation systems to ensure tasks are completed reliably and improvements proven before reaching users.

The role emphasizes end-to-end agent development, quality metrics, and scalable evaluation workflows, with a focus on applied AI and ML systems in a fast-moving team.

Qualifications

• 3+ years in data science or ML roles focused on evaluation systems, metrics, or quality measurement.
• Proficient in Python and SQL for analysis and automated pipelines.
• Experience designing evaluation frameworks with measurable success criteria and benchmarks.
• Degree from a strong engineering program; demonstrated progression in roles.
• Experience with applied AI/ML systems and collaborative problem-solving.

Responsibilities

• Build end-to-end agents with minimal executive oversight.
• Define and measure agent quality with reliable metrics.
• Develop agent capabilities for existing executive tools and evaluate readiness.
• Investigate production failures and ship fixes to prevent repeats.
• Build detection systems to surface failure patterns at scale.
• Improve agent recovery when tools or websites fail.
• Create repeatable evaluation workflows to measure impact of changes.

Skills

Data science

Machine learning

Evaluation systems

Team collaboration

Education

Bachelor's degree in engineering

Tools

Python

SQL

Django

React

TypeScript

Job description

OpenKrill in Palo Alto, CA is building autonomous AI capabilities for executive productivity. You will develop agent capabilities and evaluation systems to ensure tasks are completed reliably and improvements proven before reaching users.

The role emphasizes end-to-end agent development, quality metrics, and scalable evaluation workflows, with a focus on applied AI and ML systems in a fast-moving team.