insights
  • Home
  • Blog
  • Projects
  • Tools
    • Prompts
    • Climate Policy Dashboard
  • About
  • Work with me

Projects

bespoke tools that make a specific business process measurably better

Bespoke tools, each aimed at a single concrete friction: a model an analyst was rebuilding by hand, a diligence read that took weeks, a decision buried in scattered data, a workflow that should have been automated years ago. Every one is client work, built end to end and handed over, grouped by what it proves. More on how the work happens is on the about page.

techno-economic models

Defensible models where the assumptions have to hold up in front of a board, lender, or regulator.

first-of-a-kind cleantech valuation

Process-engineering-grade modeling and a live techno-economic valuation platform, every assumption evidence-backed, built for first-of-a-kind cleantech from methane destruction to cement decarbonization.

techno-economic process engineering valuation

AECO gas fundamentals

A natural gas supply, demand, and basis model with LNG netbacks and liquids-driven completion economics, computed live in the browser.

techno-economic energy markets python

decarbonization opportunity model

Operator-level emissions economics across hundreds of energy operators, ranking who should decarbonize first and whether it pays.

emissions economics carbon pricing data

economic impact calculator

An input-output model rebuilt from 1,500 lines of legacy R into a fast API that returns jobs, GDP, and tax impacts in seconds.

input-output policy analysis FastAPI

diligence and capital

Fast, defensible reads and investor-grade materials for technical decisions.

cleantech due diligence

An engineering-first diligence framework that catches cleantech deals breaking thermodynamics before the capital does.

technical validation diligence frameworks

climate adaptation platform

An interactive investor platform where the climate scenario drives the market, the returns, and the pipeline, with no backend.

fund modeling investor materials python

where AI actually pays off

A practical filter for finding AI work worth funding: bottlenecks before platforms, narrow measurable pilots, computable ROI.

AI strategy finance

agentic systems

Production AI agents that do real operational work and keep a person on every decision that matters: deal-making, leasing, regulatory operations, and the back office.

oil & gas operations agents

A suite of AI agents for upstream and midstream operators, covering production accounting, regulatory filing, emissions, contracts, and pipeline integrity, grounded in the public record.

AI energy data

deal copilot

A copilot for a boutique merchant bank that ingests messy company data, audits the financial model without editing it, and drafts source-traced deal materials.

AI finance systems

consultation operating system

Watches official registries and tracks every proponent commitment in a tamper-proof ledger until it is kept or breached.

AI policy auditability

private ai, installed and owned

Reads documents, routes intake, chases receivables, and reports across the tools a business already uses, running on hardware it owns.

AI systems infrastructure

ai leasing desk

Runs the full leasing lifecycle, from instant lead engagement across channels to an explainable renewal-pricing engine.

AI data systems

agent control plane

A python-first design for production agent systems where every action is policy-gated, replayable, and auditable.

AI systems architecture systems

analytics and data systems

The systems behind the decision: pipelines, simulation, and AI work without black-box risk.

document intelligence engine

Turns a mountain of messy documents into a connected map you can question in plain language, every answer cited and the quality proven, while your experts keep the decision.

AI knowledge graph human-in-the-loop

AI carbon accounting

A platform that meters the emissions of AI inference and ties offsets to billing, with signed receipts and a full audit trail.

data systems AI auditability

energy agent simulation

A multi-agent simulation framework for analyzing energy trading strategies and market dynamics, built for scale.

agent-based modeling simulation python

other work

Client builds outside the categories above.

dene language app

An offline-first flashcard app for learning a Dene language, built with spaced repetition and respect for the elders who hold it.

PWA language

Live, usable tools (Prompts, the climate policy dashboard) sit under Tools in the nav.


Looking for the analytical version of this? See how I work with teams →.

© 2025 | running on coffee & python