Research Engineer – Applied AI

HeyMilo AI

Research Engineer building reproducible reinforcement-learning environments, reward functions, and evaluation harnesses. Improving conversational AI interviewers through model evaluation and post-training experiments at HeyMilo.

Posted 8/5/2026full-timeSan Francisco • California • 🇺🇸 United StatesMid-LevelSeniorWebsite

Core Competencies

Role fit

Core Competencies

Use this summary to align your resume positioning with the role.

Demonstrates expertise in designing reinforcement learning environments and reward functions, with strong software engineering skills in Python and hands-on experience with LLMs. Capable of conducting evaluations and fine-tuning models while contributing to research and publications.

Highest-signal resume keywords

Reinforcement LearningPython ProgrammingLLM Fine-TuningDockerEvaluation Methodology

ATS Keywords

Tailor your resume

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills

Reinforcement LearningReward DesignPolicy OptimizationEvaluation MethodologyPython ProgrammingLLM Post-TrainingAgentic LoopsFine-TuningEvaluation FrameworksSimulated Environments

Soft Skills

Problem OwnershipAdaptability

Tools & Technologies

DockerLinuxCloud Environments

Certifications & Qualifications

Master's or PhD in AIMachine LearningComputer Science

Industry Keywords

Machine LearningReinforcement LearningEvaluationOpen-Weight ModelsResearch Publications

Tech Stack

Tools & technologies

CloudDockerLinuxPython

About the role

Key responsibilities & impact

Design and build reinforcement learning environments for specific real-world use cases, including realistic simulators, tool interfaces, and episodic task generation
Design verifiable reward functions that score intermediate actions and end states while resisting reward hacking
Build and maintain evaluation harnesses for task suites across frontier and open-weight models, including scoring and cost tracking
Run post-training experiments such as RLVR-style fine-tuning of open models
Package environments and results for reproducibility
Contribute to research write-ups and published evaluations
Help define future use cases based on model weaknesses
Work directly with founders and research advisors from problem definition through reproducible environments for model evaluation and training

Requirements

What you’ll need

Master's or PhD in AI, Machine Learning, Computer Science, or a closely related field
Solid grounding in reinforcement learning and LLM post-training, including reward design, policy optimization, and evaluation methodology
Strong software engineering skills in Python
Hands-on experience with LLMs, including running evaluations, building agentic loops, tool calling, and fine-tuning
Comfortable with containers and infrastructure including Docker, Linux, and cloud environments
Ability to operate in ambiguity and move quickly while owning a problem end to end
Based in or willing to relocate to the San Francisco Bay Area
Published research or open-source contributions in ML, RL, or evaluation preferred as a bonus
Experience with RL/evaluation frameworks and simulated or sandboxed environments is a bonus
Experience training or fine-tuning open-weight models at any scale is a bonus

Benefits

Comp & perks

Equity
Benefits
Opportunity to work directly with founders and experienced research advisors
Name on published work
Ground-floor role with real influence over how the company builds