In this episode of Software Leaders Uncensored, host Steve Taplin (CEO of Sonatafy Technology and author of Fail Hard, Win Big) sits down with Jeff Prince, an engineer-turned-leader and CTO at XOi, to discuss building software for field technicians, scaling engineering teams, and how AI is changing modern engineering workflows.
What XOi does and the problem it’s solving
Jeff explains that XOi is a data intelligence solution for equipment, focused on building knowledge across the full asset lifecycle, including how equipment is made, managed, and maintained. XOi works with field service companies, manufacturers, and distributors to provide actionable intelligence about equipment.
A major challenge XOi addresses is that critical field knowledge has historically lived inside individual technicians’ heads. As many experienced technicians retire, XOi aims to preserve that knowledge and make it accessible so newer technicians can make better decisions about assets.
Jeff’s path to CTO and scaling the engineering organization
Jeff joined XOi through a former colleague who was consulting with the company. XOi hired him as VP of Engineering to establish processes and build the team. At the time, the engineering group consisted of around five people with limited process and no QA function. Over time, Jeff helped grow the team to approximately 30–35 people and later transitioned into the CTO role.
Team structure and work model
XOi operates with a fully remote engineering team distributed across the United States and supported by a small number of outsourced partners. The organization is structured into squads of four to six people aligned to products or initiatives. The team includes developers, platform engineers, QA engineers, and specialists in data science and data engineering.
Designing for field technicians
Jeff highlights usability as one of the biggest challenges in building software for field technicians. These users are focused on completing physical work, not interacting with complex software. As a result, the product must deliver value without slowing them down.
Many users are historically more mechanical than software-oriented, though newer technicians are increasingly open to technology that helps them work faster and more efficiently. Regardless, simplicity remains critical.
- Clear UI and UX
- Mobile-first design
- Offline functionality
- Reliability in low-connectivity environments
Capturing field data with voice and camera
Jeff describes an AI-driven workflow currently in alpha testing that allows technicians to open the mobile app, activate the camera, and verbally describe equipment while capturing photos. The system transcribes the audio, processes the images, and converts the information into structured equipment data.
This data is enriched in the background and powers XOi’s broader intelligence platform. From the technician’s perspective, the workflow is designed to be fast: take photos, speak naturally, and move on.
Human-in-the-loop validation
To address AI accuracy concerns, XOi maintains a human-in-the-loop approach. Captured data is presented back to the user for review and correction in real time. Technicians can also validate and confirm data later through a desktop browser experience when mobile review is impractical.
AI coding tools and lessons learned
Jeff shares that XOi has seen productivity gains from AI coding tools, along with important lessons. Rapid code generation sometimes resulted in difficult-to-maintain code, requiring teams to step back and refactor. As a result, the team has refined prompts and clarified standards before generating code.
- Claude Code is the primary tool in use
- GitHub Copilot is also supported
- Other tools evaluated included Cursor, Windsurf, and Aider
The evolution of software engineering
Jeff frames AI-assisted coding as the next step in a long evolution of software development. While AI reduces the need to manage low-level concerns, engineers must still understand system architecture and be able to identify incorrect or suboptimal outputs.
Managing tech debt
XOi moved away from allocating a fixed percentage of time to tech debt and now prioritizes it alongside features in the backlog. Tech debt is evaluated based on its impact on engineering efficiency and effectiveness, while features are evaluated on business outcomes like revenue and retention.
Product and engineering alignment
To improve alignment, XOi involves engineering leaders earlier in the product lifecycle. By understanding the “why” behind initiatives, engineers can contribute more effectively and collaborate on realistic delivery timelines.
Jeff notes that meeting cadence varies by team and project complexity, and that periodic in-person working sessions have proven highly effective for complex initiatives.
Onboarding engineers
New engineers are paired with existing team members and spend their first three to four days working together in person. This accelerates environment setup, system understanding, and overall productivity. XOi also prioritizes well-known, standard technologies to reduce onboarding friction.
Key engineering metrics
- Velocity, used to identify trends rather than judge performance
- Cycle time from feature start to delivery
- Defect escape rate as a quality signal
Communicating with non-technical stakeholders
Jeff emphasizes tailoring conversations to stakeholder priorities, whether revenue, predictability, or cost. He focuses on using data to clearly explain tradeoffs and expected outcomes.
Lessons learned and advice
Jeff shares that one of his biggest lessons was learning to step back from day-to-day execution to spend time reading, learning, and thinking strategically. He closes with practical advice for tech leaders: stay informed about the industry, connect with peers, and build a trusted network.