
AI Reliability Intern
What you need
Tech
Experience
- Projects and coursework count
What you'll do
- Review agent workflow traces
- Write Python scripts for automation
- Reproduce and document issues
- Verify fixes with engineering team
We’re looking for an AI Reliability Intern to help us understand where our customer’s AI agent workflows go wrong. The work requires critical thinking and pattern matching beyond technical skills. We’re not trying to find where our agent workflows aren’t working as intended, we’re going beyond that to help our customers define better workflow agents so that they can be succesful.
You’ll:
• Review agent workflow traces and flag anomalies, bugs, and missing cases.
• Write simple Python scripts to spot patterns and automate checks.
• Reproduce issues and clearly document what happened.
• Work with the engineering team to verify fixes.
You should be curious, detail-oriented, comfortable with basic Python and data analysis, and interested in LLMs or AI agents. Prior professional experience isn’t required - projects and coursework count.
This internship is open to candidates working remotely or onsite in India, with flexible hours.
About Us
We’re a closely-knit team that believes the only thing that matters in the age of AI is making our customers succesful. We’re all hands-on engineers who are excited by making AI work for everyone, not just those at the frontier.
Interview Process
We try to keep things simple and efficient. 45 minute interview, followed by offer with 2-week probationary period.
Optimize your resume for this job
Get a match score and the keywords you're missing
About Patched
Patched executes regulated workflows in hours instead of days, with full auditability and control.
Similar Software roles


