Research Engineer: Self-Improving AI Agents

Console; San Francisco (CA); On-site; USD 150,000 - 210,000; Full time

Benefits offered by this job

Health, dental, and vision insurance

Unlimited PTO

401(k)

Equity

Meals provided daily in office

Job summary

Console is building AI agents that automate enterprise tasks with minimal human intervention. We’re hiring a Research Engineer to push our core agent loop toward self-improvement, turning production traffic into better agents through evals, labeling workflows, and targeted model adaptation.

You will work at the intersection of applied AI engineering and research, building systems to measure, debug, and improve agent behavior in real‑world conditions with production traces, offline replay, and

Qualifications

• Advanced degree or equivalent experience in building production AI systems.
• Strong software engineering fundamentals and ability to design, build, and debug complex systems.
• Experience building evals for AI systems with datasets, metrics, and offline replay.
• Knowledge of modern model adaptation methods incl. LoRA/QLoRA, SFT, distillation, reward modeling, DPO/GRPO.
• Ownership mindset to drive end-to-end projects and ship measurable improvements.

Responsibilities

• Build and improve AI systems operating in real‑world conditions with emphasis on reliability, speed, and cost.
• Systematically improve agent behavior across prompts, programs, routing, constraints, and model adaptations.
• Fine‑tune and evaluate specialist models for high‑volume tasks with production feedback.
• Collaborate with product, engineering, and leadership to ship changes into production quickly.

Skills

Software engineering

Machine learning

Applied AI

Production AI systems

System reliability

Education

Advanced degree in CS/ML

Tools

DSPy

GEPA

LoRA/QLoRA

SFT

Distillation

Job description

Console is building AI agents that automate enterprise tasks with minimal human intervention. We’re hiring a Research Engineer to push our core agent loop toward self-improvement, turning production traffic into better agents through evals, labeling workflows, and targeted model adaptation.

You will work at the intersection of applied AI engineering and research, building systems to measure, debug, and improve agent behavior in real‑world conditions with production traces, offline replay, and