AI Engineer (Self-Evolving Agent)

The role

This person will build a Self-Improvement Loop to generate, evaluate, and improve reusable skills, knowledge, strategies, and workflows from agent execution experience and feedback. They will also develop Agentic AI Architecture and Execution Runtime to design planning, tool-use, context management, and orchestration.

What you will do

• Build a Self-Improvement Loop that generates, evaluates, and improves reusable Skill, Knowledge, Strategy, and Workflow from Agent execution experience and Feedback.
• Build a Data Flywheel that connects actual service Interaction and Execution Trace to training and evaluation data and verifies improved Capability in actual vehicles.
• Design LLM-based Agent Planning, Tool-use, Context Management, and Orchestration and develop Agentic AI Architecture and Execution Runtime.
• Experiment and verify latest research for Gleo and implement it as Production Capability while responding to Safety Risk and Failure Mode.

Required qualifications

• Master's degree or higher in Computer Science, Artificial Intelligence, or related fields
• 4 or more years of practical experience in AI/ML or software engineering
• Understanding of core principles of LLM and Agentic AI and experience applying them to real services
• Ability to quickly understand latest research and verify applicability through hypothesis, experiment, and evaluation
• Software development capability to design, implement, and operate Production-level AI/ML systems
• Ownership to define problems without fixed answers and create results delivered to actual users
• Collaboration capability to discuss intensely based on evidence and align with decided directions

Nice to have

• Experience researching or implementing Self-Improvement systems where Agents learn and improve from experience and feedback
• Experience building and operating Agent Evaluation, Benchmark, LLM-as-a-Judge, and Regression Testing systems
• Experience building Data Flywheel or Continuous Learning Pipeline using large-scale Interaction and Execution Trace
• Experience researching or applying Agent systems such as Multi-Agent, MCP, A2A to actual services
• Experience responding to Agentic AI Safety and Failure Mode such as Prompt Injection, Memory/Skill Poisoning, and Tool-use Safety
• Experience measuring and optimizing LLM service Latency, Cost, Reliability, and Task Performance
• Experience publishing papers in major conferences, contributing to open source, or connecting research ideas to Production Capability
• Experience designing and developing AI/LLM services considering device characteristics such as vehicles or mobile

Before you apply

• Interview process may change depending on schedule and progress
• Exclude prohibited information such as resident registration number, family relations, marital status, salary, photo, physical conditions, and hometown from the application
• Employment may be canceled if false information is found in the application
• Reference checks may be conducted after the interview process with the applicant's consent

About the company

Gleo AI is an LLM-based vehicle voice AI agent first applied to the Hyundai Motor The New Grandeur. The Gleo Core team creates the LLM/agent systems at the center of this experience.