AI Research Scientist

Syntic AI, Inc.; United States; Remote; USD 120,000 - 180,000; Full time

Job summary

Syntic AI, Inc. is building the software runtime that enables AI agents to plan tasks, use tools, retain context, and recover from errors. This role blends applied AI research with backend engineering to make agent behavior measurable and understandable.

You will design and implement systems for coordinating agents, planning, memory, and recovery, while evaluating performance to drive improvements and safe interruption mechanisms. Join a hands-on, mission-driven team.

Qualifications

• Experience building production applications or agents powered by large language models.
• Strong backend engineering skills in TypeScript or Python.
• Understanding of how tool-using models can fail.

Responsibilities

• Design and implement systems for coordinating agents and invoking tools.
• Develop runtime capabilities for planning, memory, and recovery.
• Evaluate agent performance and use measured results to improve task completion.
• Make agent actions observable and provide ways to interrupt behavior safely.
• Account for common failure modes in models that use tools.

Skills

Backend engineering

Agent systems design

Tools

TypeScript

Python

Tool invocation systems

Job description

Role overview

Build the software runtime that enables AI agents to plan tasks, use tools, retain context, and recover from errors. The work combines applied AI research with backend engineering, with an emphasis on making agent performance measurable and behavior understandable.

Responsibilities

• Design and implement systems for coordinating agents and invoking tools.
• Develop runtime capabilities for planning, memory, and recovery.
• Evaluate agent performance and use measured results to improve task completion.
• Make agent actions observable and provide ways to interrupt behavior safely.
• Account for common failure modes in models that use tools.

Requirements

• Experience building production applications or agents powered by large language models.
• Strong backend engineering skills in TypeScript or Python.
• Understanding of how tool-using models can fail.