Mass torts create unique technical challenges. Getting 100,000 people to fill out forms or sign documents requires emails, texts, phone calls, predictive dialer; infrastructure that looks more like consumer tech than traditional legal practice. Joel's role evolved from vendor management and tool consolidation to custom software development and now commercial products that Keller Postman licenses to other firms.
The team uses GitHub Copilot extensively, and the stats look impressive. But Joel pushes back on the hype: AI writes rough drafts, not finished products. The real work, understanding requirements, making architectural decisions, finding synergies across systems, still requires human judgment. "If you view it as a first draft, that's accurate."
Rather than letting AI generate random new patterns every time, Joel's team feeds their design patterns and task templates into AI models. This creates consistency while making a process developers hate like writing out tasks in Jira or Azure DevOps more enjoyable through iteration with AI. The key: AI learns their strengths and existing patterns rather than introducing chaos.
Keller Postman receives thousands-page medical records and needs to find specific citations that legally prove injuries. The team built over 100 medical review questionnaires on the same platform they originally created for client intake forms. "The users never asked for that. They asked for specific data extraction. But we found the synergy."
Joel's background as a collegiate wrestler at University of Illinois (Big Ten champions) shaped how he approaches software leadership. Operating under pressure, limiting unknowns, deliberate practice—these aren't just athletic principles. "You can't be comfortable being uncomfortable in software engineering and then try to live differently. When you need to push through solving a problem, you're not doing it just then. You're doing it all the time."
Keller Postman has acquired multiple firms, each with different CRMs (Salesforce-based, FileVine, Clio) and varying levels of modernization. Joel discovered that while due diligence identifies tech risk, the integration challenge is always figuring out how to move people forward without overwhelming them emotionally. "It's very emotional to be acquired and have new leadership tell you these are the new tools. The pace of the plan changes drastically based on how people react."
Joel learned from a QBasic textbook at age ten. When his daughter opened that same book, none of the screens existed anymore. So she asked ChatGPT: "Teach me how to code." The ability to interact, ask what you don't understand, and focus on specific areas, rather than following step-by-step instructions, is a game changer for learning.
Stay tuned for more conversations on building tech in unexpected industries, applying athletic mindsets to engineering, and finding the balance between buying and building on CTO2CTO.