SlipstreamJobs tracks this role from the company's public career site. Apply directly on the employer's site.
Onebrief builds collaboration and AI-powered workflow software for military planning and operational coordination. The company helps military teams coordinate information, people, and decisions across systems and locations, enabling them to test strategies, adapt to changing conditions, and make decisions with greater clarity.
You will join Onebrief's Outcome Engineering team, which is inventing a new paradigm in software development. Rather than being constrained by human bandwidth, the team is building an AI Engineering Platform that transitions from managed backlogs to infinite execution through agentic systems.
In this role, you will:
- Architect multi-agent systems that enable specialized agents to collaborate, debate, and execute complex tasks reliably, moving from simple prompts to robust, stateful agentic workflows.
- Implement automated governance and policy engines that allow autonomous systems to operate safely at speed, replacing manual review with programmatic constraints.
- Engineer semantic infrastructure (knowledge graphs, vector stores) that provides agents with deep structural understanding of the codebase and business intent.
- Build rigorous evaluation frameworks that move beyond subjective validation to deterministic validation of agentic outputs using formal specifications and automated metrics.
- Design self-healing loops that connect observability data to agent actions, enabling systems to detect, analyze, and autonomously remediate failures.
- Prototype and deploy new agentic tools to production for both internal teams and customers, learning through rapid iteration.
- Experiment continuously with new models, orchestration techniques, and validation methods in this emerging field.
The team is distributed but high-bandwidth, with members working remotely and some working alongside customers in operational environments globally. Founded in 2019, Onebrief is backed by General Catalyst, Battery Ventures, Insight Partners, Sapphire Ventures, and Human Capital, and is valued at over $2 billion.
Core technologies include large language models, vector databases, Node.js, TypeScript, PostgreSQL, Redis, Kubernetes, and AWS.
REQUIREMENTS:
The company seeks pioneers from diverse backgrounds. Ideal candidates may come from:
- Software engineering: Strong coding skills, viewing code as a liability and outcomes as the asset; experience building products, tools, platforms, and force multipliers.
- AI research: Deep understanding of models and their application to production systems; expertise in emergent behavior and multi-agent coordination.
- Game engineering: Experience building complex simulations or games; understanding of rules, entities, and validation logic for autonomous agents.
- Product management: Obsessive about products; ability to translate ambiguous user needs into rigorous deliverables and manage complexity.
- Platform/SRE engineering: Infrastructure as product; expertise in observability, reliability, and automation applied to probabilistic AI systems.
- Design: Comfort creating with technology; focus on building agentic systems that enable more humane products and decisions.
- Agentic pioneers: Regardless of background, must have already built agents, hit the limits of frameworks like LangChain or AutoGen, and have strong opinions on orchestrating reliable reasoning.
Non-negotiable requirements:
- Must be a product builder with a history of shipping; understand the difference between prototypes and products users rely on.
- Must be an active agentic builder; have built with agents recently and currently; understand the frustration of prompt failures.
- Must love intense collaboration; comfortable with high-bandwidth remote work, synchronous debate, and shared problem-solving.
- Must respect expertise without gatekeeping; willing to share deep knowledge and empower others across disciplines.
- Must write clearly to think; demonstrate clarity through documentation of experiments, architectural decisions, and failures.
