I'm Paul, Agent + Tools Team Lead at SpaceXAI. I build applied AI systems and bring a scientist's approach to evidence and a teacher's approach to explanation.
Agent + Tools Team LeadSpaceXAI · 2025–present
Ph.D. scientist & educatorWSU Entomology · USDA research · Science, Python & R teaching
A small arcade makes a difficult systems problem visible: coordinate persistent agents, keep games moving through model latency, recover state, and let visitors join the conversation.
Continuous emulator clocks and live cabinet streams.
Per-resident saves, fenced control and durable message receipts.
Separate public, player, manager and lifecycle responsibilities.
Fifteen installed games. Agent availability depends on the VM worker and provider. Game completion is not verified.
Field observation, study design and analysis: connected parts of research.
02 / Scientific research
Asking better questions. Testing the answers.
My research background spans ecology, biological control and experimental design. The habit carries into engineering: does an evaluation actually measure what it claims?
Design, build, test, explain: a practical approach to learning.
03 / Teaching
Understanding through building.
I've taught science and programming, from high-school labs to college courses. A lesson becomes more useful when learners can test an idea and build something with it.
Inspect synthetic inputs, learned weights and held-out measurements from a real local calculation. This small experiment makes the computation visible.
I'm a software engineer and a Ph.D. entomologist. Research, teaching and building software have each taught me to look closely at a problem, make the assumptions visible, and learn from what happens.
This portfolio brings those parts of my work together. The technical examples here are original public experiments.