Job Description
We’re looking for an AI engineer. You’ll work on features that help developers understand their agents’ behavior, evaluate their performance, and improve them over time.
This is a hands-on engineering role. You’ll move between experiments, benchmarks, and product code, taking what you learn from evaluating agents and using it to shape the framework.
Why
- We’re starting to build more and more agentic features into Mastra: Trace intelligence, and we’re working on various Agent Learning features to automatically self-improve agents (Autoevals + Suggestions, Rollouts).
- In addition, we’ve been increasingly running Mastra primitives like memory on industry-standard benchmarks and using the results to fine-tune configuration.
What you’ll do
Build the features that make this possible: tools for understanding traces, evaluating agent behavior, and turning results into improvements.
Run experiments on Mastra’s primitives, investigate where they perform well and where they fall short, and use those findings to improve their configuration and behavior. You’ll help turn that work into capabilities developers can use in their own applications.
You have
- Shipped agents and agentic applications into production
- Tuned and tweaked them to improve accuracy
- You need to be able to benchmark your results against existing datasets, or intelligently creating.
About Mastra
Mastra is the open-source TypeScript framework for building AI agents. Built by the team behind Gatsby, backed by Y Combinator and Spark Capital, with $35M raised total. We went from 0 to 27k+ stars and 1M+ weekly npm downloads in under 18 months.
Small team, shipping fast, fully remote. Most of us work across North American and European time zones.
How to apply
The Apply button opens Mastra's own application page in a new browser tab — applying is free, and this board never sits between you and the employer: no middleman, no form of ours to fill first.
This posting runs on Ashby, the applicant tracking system behind Mastra's job board — the form you'll see lives at jobs.ashbyhq.com. It asks for the basics first: your name, email, and a résumé/CV upload. After that come the custom questions this employer wrote for the role — Ashby lets each team tailor its application, and some postings add a short questionnaire on top, so set aside a few quiet minutes rather than a thirty-second sprint. Because the questions change per role, re-read the requirements above before you start typing; answers that echo the exact skills the listing names are the ones that survive screening. Two practical notes: apply promptly while the posting is verified live, since teams close applications once a role fills, and use one email address for the application so your details stay under one identity.
Before you start
Quick answers
- Is this job still open?
- Yes. Listings on this board are re-verified 4 times a day and auto-expire 45 days after posting — if this page is visible, the role was open at the last check.
- Where does the Apply button go?
- Straight to Mastra's own application page — never a middleman. Every listing here comes from the employer's own job board, so the Apply button simply opens their page in a new tab.
- Is the salary listed?
- No — this employer didn't publish a salary range for this role. Their own application page is the place to check for pay details.
About the job
Skills
More Machine Learning & AI roles
View all
Staff Software Engineer, AI Foundations (AI Agent Optimization)
AI & Automation Intern – Finance
Senior Applied AI Engineer
AI Lead
AI Partnerships Manager
