ECMWF, GFS, HRRR, RRFS, GraphCast, Pangu, and many more models are producing better forecasts than ever. The organizations winning on weather aren't doing it with better data — they're doing it with a confidence intelligence layer that tells them when to trust the forecast, how to read model disagreement, and how to turn uncertainty into operational decisions.
What I Do
The era of single-model thinking is over. I help you navigate the full stack — from ensemble spread to executive decision — and build products that make uncertainty work for you.
I apply meteorological principles of ensemble spread, model agreement, and conditional skill to extract signal from the noise of competing model outputs.
From pre-seed to Series A, I help weather tech founders translate cool technology into something customers understand, need, and buy.
I bridge the gap between meteorological credibility and commercial traction — positioning, messaging, and sales enablement for a technically complex product.
I help non-weather organizations build internal frameworks for acting under meteorological uncertainty — from supply chain to financial risk.
I embed in your team to scope, build, and validate weather-intelligent features — from data architecture to UX.
Need a credible voice in the room? I advise leadership teams as a technical sounding board at the intersection of weather science and business strategy.
The Thesis
Access to ECMWF, GFS, HRRR, RRFS, GraphCast, and a rapidly expanding field of other forecast models is no longer a competitive advantage. The infrastructure barrier is gone. Everyone has the data.
When ECMWF, GFS, HRRR, RRFS, GraphCast, and many other models all say something different, most teams freeze, average blindly, or default to whoever sounds most confident. None of those is a decision framework.
The foundations of forecast confidence — ensemble spread, historical skill, conditional model accuracy, and cross-model agreement — exist precisely to answer the question: how much should I trust this forecast right now? That is the edge.
You have a technology and a thesis. You need help turning it into a product someone will pay for and a story someone will fund.
Your product works. Now you need to scale distribution, sharpen positioning, or navigate a new vertical.
Operations, finance, or strategy teams with significant weather exposure but no meteorological expertise in-house.
If you're working on something compelling at the intersection of weather science and real-world decisions, I'm interested.
Step 1: Reach Out
Send a message through the contact page. No pitch deck required — just tell me what you're working on.
Step 2: Discovery Call
We spend 30–45 minutes mapping your problem, your data landscape, and what success looks like for you.
Step 3: Scoping
If there's a fit, I put together a focused proposal: objectives, approach, timeline, and what you'll walk away with.
Step 4: Engagement
We work together — hands-on or advisory, depending on what you need. Focused, time-boxed, results-oriented.
Ready!
Use Cases
Weather intelligence isn't one-size-fits-all. Here's where the forecast confidence framework makes the biggest impact.
Applying model spread and confidence signals to weather-based market instruments, where small probability edges matter most.
Helping municipalities and contractors plan based on model agreement and ensemble spread — not a single deterministic forecast.
Demand forecasting, grid management, and renewable integration under uncertainty — where the tails matter as much as the mean.
Weather exposure in agricultural, energy, and freight markets quantified through ensemble-based risk frameworks.
Translating ensemble spread into actionable go/no-go decisions for time-sensitive outdoor activities and trip planning.
Designing confidence-aware triggers and pricing structures for weather-linked financial products.
Proactive weather-risk management in transportation using model agreement as an early warning signal.
Helping startups communicate uncertainty honestly and usefully — building products that earn trust by being right about what they don't know.
Bridging meteorological research and applied business context for institutions at the frontier of forecast development.
Deep dives on uncertainty interpretation, forecast confidence, and what the model explosion means for real-world decisions. Plus live commentary on major weather events and market moves.
The business model for private weather companies is changing. For decades the pitch was "we have the most accurate data." That was defensible when forecast quality was scarce. That scarcity is gone — and the new edge is something most providers aren't even measuring yet.
FAQs
I help weather startups build better products and go-to-market strategies, and help non-weather organizations make smarter decisions under meteorological uncertainty. The common thread is applying forecast confidence science to turn data overload into clear action.
Weather tech founders at pre-seed to Series A, operations and strategy teams with significant weather exposure, and anyone building products or decisions around meteorological data. If you're not sure if you fit, just reach out — I'll tell you honestly.
It depends on what you need. I take on hands-on product work, go-to-market advisory, and pure consulting engagements. See the Services page for more detail, or reach out to discuss your specific situation.
It's in development. The initial focus is prediction markets, snow removal operations, and energy/utilities. I'll be writing about it on the blog as it takes shape — and eventually it'll live here on the site.
Use the contact form or email me directly. No pitch deck, no formal intake — just tell me what you're working on and we'll go from there.
Because no model is always best. Skill is conditional — GFS outperforms ECMWF in certain regimes, HRRR and RRFS excel at convective timing and short-range mesoscale detail, GraphCast leads at certain medium-range leads — and many other models each carry their own conditional skill profile. Forecast confidence frameworks let you weight and combine models systematically, extracting more signal from the same data.
Book a discovery call and explore what better uncertainty interpretation could unlock for your organization.
