Frontier Agents Engineer (Applied AI)

Space Capital is a company focused on building and deploying reliable production AI applications. They are seeking a Frontier Agent Engineer (Applied AI) to bridge the gap between AI research and production deployment, working with enterprise customers to design and deploy intelligent systems that leverage cutting-edge AI technologies.

Responsibilities:

• Design and deploy production AI agents that leverage the latest advances in large language models, reasoning, retrieval, memory, and tool use
• Architect intelligent systems that combine LLMs, traditional machine learning, structured knowledge, enterprise data, and deterministic software into reliable production workflows
• Engineer customer intelligence layers, retrieval pipelines, memory systems, and knowledge representations that allow agents to reason over large, heterogeneous enterprise data
• Develop multi-agent systems that coordinate reasoning, planning, tool execution, and human oversight
• Translate frontier AI research into production systems by rapidly evaluating new models, prompting techniques, reasoning paradigms, and agent architectures
• Own the full experimentation lifecycle, from hypothesis generation to production rollout
• Design rigorous evaluation frameworks using offline benchmarks, online A/B experiments, golden datasets, regression suites, LLM-as-a-Judge, and human evaluation
• Run controlled experiments and ablation studies to understand the contribution of different models, prompts, retrieval strategies, reasoning techniques, memory systems, and agent architectures
• Continuously evaluate newly released frontier models and determine where they meaningfully improve quality, latency, reliability, or cost
• Develop confidence estimation, reflection, and continuous learning systems that improve agents over time using real-world feedback
• Measure success through business outcomes, not benchmark scores
• Build production-quality AI systems with a strong emphasis on reliability, observability, latency, safety, and cost
• Design agent guardrails, fallback strategies, tracing, monitoring, and evaluation pipelines that enable safe deployment in high-stakes environments
• Collaborate with infrastructure engineers to deploy AI systems securely within enterprise cloud environments
• Build human-in-the-loop workflows that effectively combine AI automation with expert oversight
• Partner directly with enterprise customers to understand their business, data, and operational challenges
• Translate ambiguous customer problems into production AI architectures
• Rapidly prototype new ideas, validate them with customers, and evolve successful solutions into scalable production systems
• Identify reusable patterns that become core capabilities across many enterprise deployments