AI Engineer at Mastra

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 ycombinator.com and Spark Capital sparkcapital.com, with $35M raised total. We went from 0 to 28k+ stars github.com and 2M+ weekly npm downloads npmtrends.com in less than 2 years. Small team, shipping fast, fully remote. Most of us work across North American and European time zones.

Requirements

Must-haves

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.
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.
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.

Nice-to-haves

None flagged.

Application prep

Use the role’s language and application details while tailoring your materials.

Keywords to mirror

aitypescript

Platform: ashby
Effort: quick