Director, Fleet AI Engineering

A Day in the Life:

The Director, Fleet AI Engineering will define, build, and scale AI/ML solutions that drive significant business impact. You’ll serve as a hands-on technical leader (player/coach), operating as a key individual contributor while leading a small team and partnering closely across product, data science, and engineering. This role requires strong technical understanding of modern AI systems (including how to debug and validate them) and a bias toward leveraging AI-assisted development workflows to accelerate delivery—especially for simulation-based optimization and real-time decisioning problems.

The preferred candidate will sit at our corporate office in Atlanta, however, we are open to candidates located outside of the area to work remotely.

The target starting rate is $130K; actual salary will be determined based on experience.

What You’ll Do:

• Solution Delivery: Own the end-to-end ML system lifecycle—from problem framing and offline evaluation to production deployment with observability, drift monitoring, controlled rollouts, and iteration based on real-world performance.
• Simulation-Based Decisioning: Develop simulation environments to train and evaluate sequential decision-making methods (including reinforcement learning where appropriate), focusing on simulator fidelity, reward design, offline/online evaluation, and safe real-time deployment.
• AI-Native Engineering: Build an AI-assisted development workflow (prompting and code-review standards, automated testing, documentation) and drive adoption across the team (e.g., Claude Code, Cursor, GitHub Copilot, Databricks Genie).
• Architectural Guidance: Design and scale production AI systems (data pipelines, model training/serving, feature and experiment infrastructure), balancing latency, reliability, cost, and integration into existing platforms.
• Cross-functional Collaboration: Translate business goals into measurable ML objectives, define acceptance criteria, and run experiments to validate impact.
• Strategic Leadership: Define a practical AI/ML engineering roadmap for fleet use cases, iterate quickly based on results, and mentor a small team in a player/coach model.
• AI-Native Engineering: Build an AI-assisted development workflow (prompting and code-review standards, automated testing, documentation) and drive adoption across the team (e.g., Claude Code, Cursor, GitHub Copilot, Databricks Genie).
• Architectural Guidance: Design and scale production AI systems (data pipelines, model training/serving, feature and experiment infrastructure), balancing latency, reliability, cost, and integration into existing platforms.
• Cross-functional Collaboration: Translate business goals into measurable ML objectives, define acceptance criteria, and run experiments to validate impact.
• Strategic Leadership: Define a practical AI/ML engineering roadmap for fleet use cases, iterate quickly based on results, and mentor a small team in a player/coach model.

What We’re Looking For:

• Curiosity & Craft: You genuinely enjoy “geeking out” on AI—staying current on new ideas, trying things out, and bringing thoughtful suggestions back to the team.
• Education: Bachelor’s degree in Computer Science, Engineering, Data Science, or a related field—or equivalent practical experience.
• Experience:
  • Shipped production AI/ML systems end-to-end (prototype → deployment → iteration).
  • Hands-on engineering: able to dive into code, debug model behavior, and improve performance.
  • Led small teams or technical initiatives as a player/coach.
• Technical Skills:
  • ML Systems & Modeling: Solid grasp of supervised/unsupervised learning and reinforcement learning / sequential decision-making; able to diagnose training/inference issues. Experience with LLMs/prompting/fine-tuning is a plus.
  • Python & Frameworks: Strong Python plus PyTorch/TensorFlow/Scikit-learn; strong code review practices; able to debug and harden AI-generated code. Nice to have: RL tooling (e.g., Stable-Baselines3, Ray RLlib, Gymnasium/PettingZoo) and LangChain/LlamaIndex.
  • MLOps / Platform: CI/CD for models, versionin