{
  "name": "Jason Draper",
  "url": "https://vuca.dev",
  "location": "Utah, USA",
  "lead": "I build AI agent systems for work that people stake their reputation on.",
  "roles": [
    "AI Product Engineer",
    "Forward Deployed Engineer",
    "AI Architect",
    "Technical Co-Founder",
    "Founding Engineer"
  ],
  "about": [
    "I'm a two-time technical co-founder, and a forward-deployed engineer by instinct. My method is the same in every domain: learn the work from the people doing it, then act as their technical partner to build the software they come to rely on. Most recently that domain is financial audit, a field that resists casual understanding, where I've grown accustomed to spending roughly a third of each day with the auditors who use our platform.",
    "Every product I've built from enterprise to small businesses has involved discovery in the customer's domain, implementation against their environment and constraints, delivery measured by their outcomes and adoption, while bridging the technical and communication gaps throughout the process.",
    "The agents I build operate where a wrong answer costs professional reputations. That constraint drives engineering discipline: holding accuracy steady over long contexts, making agent decisions inspectable, and guarding what deserves a human's attention."
  ],
  "vuca": {
    "title": "Why vuca.dev",
    "copy": [
      "VUCA is the U.S. Army War College's name for an operating environment: volatile, uncertain, complex, ambiguous. It was coined for strategy after the Cold War. It also describes software development as an industry, architecture design as a practice, and applied AI in particular.",
      "Models improve in depth and breadth of capabilities. Development requirements and best practices do not hold still. Agent output can be fluent and still be wrong. System design demands an elegant fusion of deterministic and probabilistic components.",
      "These are exactly the kinds of environments I enjoy."
    ]
  },
  "background": [
    "Technical Co-Founder, Punchcard — AI-powered audit automation (YC S23; backed by Y Combinator, General Catalyst, and Dell Technologies Capital)",
    "Co-Founder & CTO, FairQuote (quoting and lead-generation for local service businesses)",
    "U.S. Air Force — nuclear and missile operations officer; led teams responsible for up to 200 ICBMs"
  ],
  "selected_work": {
    "punchcard": {
      "url": "https://punchcard.com",
      "backers": "Backed by Y Combinator, General Catalyst, and Dell Technologies Capital",
      "items": [
        {
          "name": "Financial Statement Review",
          "note": "Punchcard's flagship agent. It reviews financial statements inside live audits, running thousands of independent checks per document, where auditors sign their names to the result. I led it from discovery to release and rebuilt it four times to clear a 95% accuracy bar, each rewrite driven by one question: what is the least context this agent needs to reach a decision it can defend? It became the fastest-adopted product on the platform, and the pipelines behind it have processed more than 8 million documents in production.",
          "url": "https://vuca.dev/writing/why-llms-struggle-with-financial-documents"
        },
        {
          "name": "CoAudit",
          "note": "The platform's general-purpose agent. Auditors speak to it in their own vocabulary and it translates into the platform's, routing across 80+ tools that surface only when they're relevant. It replaced a workflow-builder interface users found too abstract and cut time-to-output by 84%."
        },
        {
          "name": "Creed",
          "note": "An autonomous engineering agent that grew into a working member of the team. It finds, triages, fixes, and merges its own code behind adversarial review and deterministic safety gates: around 30 merged pull requests a week, most with no human in the loop, at a revert rate under 1%. Watching it take on real responsibility changed how quickly I think this shift is arriving.",
          "url": "https://vuca.dev/systems/sustainable-software-factories"
        },
        {
          "name": "Agent Traces",
          "note": "Our largest customer, a 2,700-person audit firm, stalled in pilot because auditors were hesitant to sign off on output they couldn't defend. I shipped reasoning traces and a review surface that made every conclusion inspectable. They moved from pilot to an annual contract and completed audits 60% faster."
        }
      ]
    },
    "earlier": [
      {
        "name": "FairQuote",
        "note": "Quoting and lead-generation software for local service businesses, which I co-founded and led engineering for. I trained a YOLOv5 model against mapping imagery to identify dying residential trees, generating 5,500+ qualified leads worth an estimated $8.2M, and replaced multi-day manual quoting with an instant, self-serve platform that processed $2M+ in quoted jobs.",
        "url": "https://thefairquote.com"
      },
      {
        "name": "U.S. Air Force",
        "note": "Nuclear and missile operations officer. I led cross-functional teams responsible for up to 200 ICBMs across 23,500 square miles and built an internal planning tool adopted by three squadrons and 360+ members."
      }
    ]
  },
  "expertise": [
    "AI product engineering",
    "forward deployed engineering",
    "AI architecture",
    "AI agent orchestration",
    "LLM reliability engineering",
    "audit automation",
    "software factories",
    "regulated-industry AI"
  ],
  "systems_built": [
    {
      "name": "Creed",
      "tagline": "Sustainable Software Factories",
      "desc": "A QA-focused factory that proactively and reactively identifies product bugs, then investigates, fixes, reviews, and approves its own code.",
      "url": "https://vuca.dev/systems/sustainable-software-factories"
    }
  ],
  "writing": [
    {
      "title": "Financial Agents: The Company That Built Excel Is Not the Best at Excel?",
      "url": "https://vuca.dev/writing/excel-agent-architecture",
      "description": "Microsoft owns the format, the app, the models, and the telemetry — and still lost Excel-agent benchmarks. The reasons distill down to two fundamentals in agent design.",
      "published": "2026-07-28",
      "tags": [
        "AI agents",
        "agent design",
        "audit automation",
        "LLMs"
      ]
    },
    {
      "title": "Albert Einstein Walks Into a Library",
      "url": "https://vuca.dev/writing/why-llms-struggle-with-financial-documents",
      "description": "Why large language models struggle with complex document analysis, told through an analogy involving a library and an army of Einsteins.",
      "published": "2025-12-28",
      "tags": [
        "LLMs",
        "audit automation",
        "AI reliability",
        "agents"
      ]
    }
  ],
  "contact": {
    "linkedin": "https://linkedin.com/in/jasontdraper"
  }
}