Machine Learning Engineer

Syntic AI, Inc.; United States; Remote; USD 120,000 - 160,000; Full time

Job summary

Syntic AI, Inc. is seeking an experienced backend-focused engineer to build runtime systems that help AI agents plan, call tools, maintain memory, and recover from errors.

This role blends backend development with applied work on agent performance, evaluation, and safe operation. You will develop orchestration systems for agents and their tool calls, implement planning/memory/recovery capabilities, and use measurable evaluations to improve task success rates.

Qualifications

• Experience building production systems that use large language models or agents.
• Strong backend engineering skills in TypeScript or Python.
• Understanding of failure modes in tool-using models.

Responsibilities

• Develop orchestration systems for agents and their tool calls.
• Implement capabilities that support planning, memory, and recovery.
• Use measurable evaluations to identify and improve task success rates.
• Make agent behavior observable and interruptible.
• Design with the failure modes of tool-using models in mind.

Skills

Large language model / agent systems

Backend engineering

Failure modes understanding

Tools

TypeScript

Python

Job description

Role overview

Build the runtime systems that help AI agents plan, call tools, maintain memory, and recover when tasks go wrong. The role combines backend development with applied work on agent performance, evaluation, and safe operation.

Responsibilities

• Develop orchestration systems for agents and their tool calls.
• Implement capabilities that support planning, memory, and recovery.
• Use measurable evaluations to identify and improve task success rates.
• Make agent behavior observable and interruptible.
• Design with the failure modes of tool-using models in mind.

Requirements

• Experience building production systems that use large language models or agents.
• Strong backend engineering skills in TypeScript or Python.
• Understanding of failure modes in tool-using models.