I learned systems by keeping them running.
Twelve years on industrial floors came first. Medium-voltage switchgear at Schneider Electric. Hyperscale data center operations at Microsoft and T5 Data Centers. Commissioning, energizing, troubleshooting, and the long stretches of maintenance nobody writes about. NFPA 70E, OSHA 10, and the habit those certifications leave in you: assume the thing is live until you have proved otherwise, with your own hands.
That job taught one lesson over and over. A drawing is a promise. The system never signed it. What is on paper and what is in the room diverge quietly, and the gap between them is where people get hurt. Everything I build now is aimed at that gap.
I build Applied Intelligence systems that show their work. A retrieval system hands back the page it read. A record can be checked against what actually happened. An agent that isn't sure is allowed to say so and hand the decision back. That's not caution. It's what you learn from being the person who signs for the outcome.
The lab is Celaya Solutions Research. It publishes instruments, not products: things built to make a question observable. Fourteen of them are listed here with their evidence, live URLs and source where there is source. What I'm thinking through as I build goes in the journal.
Mechatronics and software development meet in the same place here: a sensor, a record, a model, and the person responsible for the result. The lab runs original AI research and experimentation, then publishes the working instrument, its current state, and the evidence a reader can inspect.
Scientific provenance has a narrow meaning in this work. The source, version, method, result, limitation, and human review stay attached. A provenance chain does not make a claim true. It makes the path to that claim visible enough for another person to check.
All of it happens in El Paso, Texas, on the border, in English and Spanish. Not because I ended up here. Because the physical systems this whole industry runs on are being built here, and family is here, and I'd rather be near both. More on why the location is the point.
There's a pacemaker in my chest. It streams a heartbeat to a public record every sixty seconds, which you can watch on this site right now. It is the plainest version of the argument I keep making: put the human in the record, and let anyone check.
Industrial maintenance
2013 onwardFailure awareness.
Twelve years on industrial floors. A machine tells you it is about to fail if you know how to listen, and the cost of not listening is measured in dollars and in people. Everything since has been built for the moment a system fails.
Data-center operations
Microsoft, T5 Data CentersUptime, logging, reversible changes.
Hyperscale rooms where a change is either logged and reversible or it does not happen. Uptime is not a feature, it is the floor. That is where the habit of writing every reading down came from.
Switchgear manufacturing
Schneider ElectricQuality gates and consequence-aware review.
Medium-voltage switchgear does not get a second draft. Every unit passes a gate, and a person signs it. The review is proportional to the consequence. That is the shape of every guardrail in the instruments.
Music production
C-Cell, 2015 onwardTiming, structure, human perception.
Producing records taught structure at the scale of seconds and what attention does under a beat. A mix is an experiment in perception, and a pulse-driven game is the same experiment with a different instrument.
Applied Intelligence development
2020 onwardOperating instincts became software architecture.
The recordings on the video page are this period: chat platforms, image graphs, a mock interviewer that grades you back. The instincts from the floor turned into architecture: log everything, cite everything, keep a person accountable.
Independent research
Celaya Solutions Research, 2024 onwardThe threads, unified into instruments.
The lab is where the threads meet. Local-first, provenance-aware systems for legal, archival and industrial work. No investors. Built in El Paso, on purpose, and in public so the method is legible.
Questions that come up
- Who is Christopher Celaya?
- An independent Applied Intelligence researcher and industrial systems operator based in El Paso, Texas. He spent twelve years in critical facilities work, commissioning and maintaining hyperscale data centers and medium-voltage switchgear for Microsoft, T5 Data Centers and Schneider Electric, and now runs Celaya Solutions Research, an independent lab building provenance-aware Applied Intelligence systems.
- What does Celaya Solutions Research do?
- It builds research instruments rather than products: local-first, provenance-aware systems for industrial, archival and civic work. Each one is built to make a question about evidence handling, agent coordination or human judgment observable. The lab is at celayasolutions.com.
- What is he an expert in?
- Hyperscale data center commissioning. Data center maintenance and operations. Medium-voltage switchgear. Critical facility electrical systems. Industrial mechatronics. Industrial controls and SCADA. Building automation and HVAC control. NFPA 70E electrical safety. Industrial telemetry. Applied Intelligence. AI research and experimentation. Large language model architecture. Provenance-aware retrieval augmented generation. Multi-agent orchestration. Local-first Applied Intelligence systems. Fine-tuning and model adaptation. Archival intelligence. Software development. Applied Intelligence engineering in El Paso, Texas. Applied Intelligence for the Paso del Norte borderland.
- Is he available for work?
- For the right collaboration and occasionally the right consulting engagement, yes. He is not looking for employment. The fastest way in is an email to hello@celayasolutions.com describing what you have and what you want it to answer.