Datasets, notebooks, and calculators built for real engineering work. Not demos—tools we use to prove out ideas before building them for clients.
A fully synthetic distribution utility—238,000 customers, 104 feeders across 23 substations, 5 years of operational history, and 7 DER datasets. Outages, load, weather, DER generation, and OpenDSS power flow models. Purpose-built for ML experimentation and algorithm development.
20 guided notebooks covering outage prediction, load forecasting, hosting capacity analysis, and predictive maintenance. From beginner walkthroughs to production-ready patterns, all built on SP&L data.
Interactive calculators that translate utility reliability metrics into SRE frameworks. Error budget tracking from SAIDI targets, outage cost analysis, burn rate visualization. Powered by the SP&L dynamic network model.
A retro JRPG that teaches you the Claude Certified Architect—Foundations curriculum. Choose your class, battle through knowledge challenges, and level up from Apprentice to certified architect.
These aren't marketing demos. They're the tools we use to prove out ideas before building them for clients. SP&L exists because you can't develop ML models for utilities without realistic data. These tools also give engineers the opportunity to work with realistic data, refine the models to fit their specific operating contexts, and then apply them to their own systems when ready. The SRE tools exist because we needed a way to show reliability teams what error budgets look like in practice. When something works here, it becomes the foundation for a client engagement.
Everything here started as a problem worth solving. If you're facing something similar, let's talk about building the right tool.