AI Engineer

What you'll do

• Design agent cognitive architectures for sustained planning, reasoning, memory, and decision-making; build scalable multi-agent orchestration.
• Improve frontier models through post-training, fine-tuning, and scaffolding; develop evaluation frameworks for reliability and regression detection.
• Turn live operational feedback into training data and benchmarks, and partner with forward-deployed and platform engineers to deploy improvements.

What they ask for

• Hands-on experience building agentic AI systems that operate autonomously over extended periods.
• Deep familiarity with frontier models; experience fine-tuning open-weight LLMs, applying post-training or architectural changes, and shipping results.
• Strong software engineering and systems skills, including building reliable, observable, debuggable production systems; Python and PyTorch.
• Data engineering rigor around provenance, labeling, and evaluation, plus working knowledge of quantization, GPU memory management, and inference optimization.

Nice to have

• Experience with agents in adversarial environments, multimodal agentic systems, or air-gapped or classified deployments.
• Active, recent, or eligibility for a U.S. security clearance; TS preferred.

Tools

Python, PyTorch, SFT, preference optimization, LoRA/QLoRA

AI focus

Develop long-horizon autonomous agents and multi-agent systems; adapt frontier models through post-training and domain-specific fine-tuning, and evaluate and improve performance using operational data.