Outcome Engineer - Experienced Professional

About the role

Onebrief is building a new discipline called Outcome Engineering, where software is created by autonomous systems guided by clear intent rather than manual step-by-step coding. In this role, you will own the design and execution of AI-powered workflows that replace traditional software delivery pipelines with systems that operate at the speed of intent. You will be responsible for architecting multi-agent orchestration layers that enable specialized agents to collaborate, debate, and execute complex planning and operational tasks reliably. You will implement automated governance and policy engines that ensure autonomous systems operate safely, securely, and within defined constraints without human bottlenecking. You will engineer context and memory infrastructure, including semantic representations and knowledge graphs, so agents maintain deep structural understanding of mission, code, and intent. You will build rigorous evaluation frameworks that validate agent outputs against formal specifications, moving decision-making from intuition and vibes to measurable, auditable outcomes. You will prototype and deploy agentic tools rapidly, proving concepts in production environments for both internal teams and external customers who rely on mission-critical clarity and coordination. This role demands pioneers who are comfortable in a field where the playbook is still being written and who are motivated by consequential work where decisions directly affect real-world operations.

Key facts

What you'll do

• Architect Multi-Agent Systems by designing orchestration layers that allow multiple specialized agents to collaborate, debate, and execute complex tasks reliably over long-running workflows.
• Implement Automated Governance by creating guardrails and policy engines that enable autonomous systems to operate safely at speed while enforcing compliance, security, and organizational constraints.
• Engineer Context and Memory by building semantic infrastructure such as knowledge graphs and vector stores that provide agents with deep, structural understanding of code, plans, and business intent.
• Build Rigorous Evaluation Frameworks by developing testing harnesses that define success through deterministic validation of agentic outputs using formal specifications and automated metrics.
• Design Self-Healing Loops by connecting observability data directly to agent actions so systems can detect failures, autonomously analyze root causes, and execute remediation without human intervention.
• Prototype and Deploy agentic tools rapidly, shipping production-grade capabilities for both internal teams and customers, with a focus on learning by shipping and measuring real operational impact.
• Experiment and Learn continuously by exploring new models, orchestration techniques, and validation strategies to understand emergent behavior in multi-agent systems and refine outcomes.
• Collaborate with cross-functional stakeholders including military planners, operational experts, and product teams to translate ambiguous, high-stakes needs into rigorous, executable software specifications.
• Contribute to the definition of the AI Engineering Platform by setting standards for reliability, safety, performance, and developer experience in an agentic world.
• Mentor and enable other engineers by documenting patterns, building internal tools, and demonstrating best practices for outcome-focused development in high-consequence environments.
• Drive clarity in product requirements by working closely with leadership to turn strategic objectives into measurable outcomes that autonomous systems can execute.
• Maintain a bias for action by prioritizing shipped experiments over theoretical perfection, using real-world feedback to guide the evolution of the platform.

Requirements

You are a coder who is ready to automate your own job and view code as a liability while treating outcomes as the primary asset that the organization is paid to deliver. You possess strong software engineering fundamentals but see software not as a collection of features but as a mechanism for reliably executing intent in complex, dynamic environments. You understand that the era of human bandwidth as the main constraint on delivery is ending, and you are excited to build in a world where AI provides the scale and adaptability previously impossible. You have an interdisciplinary background or are willing to collaborate across disciplines, recognizing that the best solutions often emerge from diverse perspectives and ego-free collaboration. You are comfortable working alongside customers in operational environments and understand the weight of decisions that flow from the systems you help build. You embrace rigorous thinking and are fluent in translating ambiguous, high-stakes problems into formal specifications that can be validated by automated systems. You value reliability, safety, and clarity over novelty, and you are prepared to implement governance and observability as first-class design requirements. You are a pioneer willing to experiment with new models, orchestration techniques, and validation strategies even when the results are uncertain, documenting learnings to accelerate collective progress.

Nice to have

Experience with multi-agent systems, emergent behavior, and advanced orchestration patterns is preferred. Background in AI research, game engineering, simulation, or platform/SRE roles that emphasize observability and reliability is especially valuable. Experience in defense, intelligence, or mission-critical planning environments is relevant but not required.

Practical notes

This is a full-time position based in the United States with remote flexibility. The role may require travel to operational environments and compliance with U.S. visa sponsorship requirements as applicable.