The founder / JCEE Labs
Jonathan Matthew Chadbourne
Independent researcher and solo founder of JCEE Labs in Dallas, Texas. I build software around a practical question: what would someone need to establish what actually happened?
A personal introduction
The person behind the work.
My background is in sales. My attention is on the practical work around software: the handoff that needs another check, the exception someone has to investigate, and the record that helps the next person understand a decision.
I started JCEE Labs to pursue questions that kept leading back to evidence. A system can report success, and an AI can give a convincing explanation. I want to know what supports that account, what remains uncertain, and who is responsible for the next step. As a solo founder, I direct the research and development you see here and take responsibility for its claims.
The starting point
From reflection to accountable action.
Mirrored was my first project. It brings together the ideas of voice conversation, daily check-ins, journaling, philosophy, and habits in a tool for reflection. It remains in development.
That starting point is personal: how can a conversation help someone notice patterns, think more clearly, and decide what to do next? The wider work asks related questions about software. What did it observe? What did it infer? What did it actually change? When software begins taking action, those distinctions affect the people relying on it.
JCEE Labs developed as those questions moved into research and practical operations. The connection is still straightforward: preserve what supports a conclusion, make uncertainty visible, and keep human judgment involved when the evidence runs out.
Current work / October 19, 2026
Making interrupted work understandable.
My current focus is JCEE-INFRA: connecting permission, actions, and recovery across software systems. An approved request can reach a database and lose its response. Permission can change while work is pending. The person supervising that workflow needs to know whether an action happened and whether another attempt is allowed.
On October 19, a tested software candidate moved into working integration with the tested source preserved. In a controlled PostgreSQL ledger using fictional data, authorized actions completed, recovery accounted for an existing effect without duplicating it, and an uncertain outcome stayed held for human resolution.
This result has specific conditions, including trusted services and a separately protected checkpoint that survives the tested restoration. Production admission, physical commitment, and independent human review remain open. The public record explains what the tests establish and what still needs work.
How I work
Ask a clear question. Keep the result inspectable.
I begin with a concrete problem and define what a useful result would show. Then I examine failures as well as successful runs: missing information, changed permission, interrupted processes, and records that disagree. When a test exposes a problem, the failure belongs in the record alongside the repair.
AI tools assist my development and review. I direct the questions, choose the next steps, and remain responsible for the decisions. Tests, retained records, and the conditions actually examined establish what I can claim. I identify AI-assisted review separately from independent human review.
For business software, that same approach begins with listening to the people doing the work. A useful improvement needs to fit their workflow and make the next decision easier to understand.
Selected engineering notes
Start with the work itself.
An open invitation
Talk with me directly.
Bring a workflow that is difficult to trust, an interruption that is difficult to recover from, or a research result you want to examine. We can discuss the problem, the evidence that would make progress visible, and whether I can help. You can request a consultation or reach me through the JCEE Labs email below.