We build the software and AI systems that make sustainability strategy operational.

Many sustainability plans stall because nothing is in place to run them day to day. Nearly two decades of this work taught us to build that part too: custom software and AI that turn a sustainability plan into tools your team uses every day.

Where CF Labs comes from

For nearly two decades, Council Fire has helped organizations build sustainability strategies and carry them out. Council Fire Labs builds on that experience, which is why we can usually see where a plan needs support before it stalls.

  1. 01

    Nearly two decades of strategy and implementation.

    We write strategies, lead the engagement that tests them, and work with the boards that fund them, for companies, governments, NGOs, and foundations. That experience shows us early where a plan is likely to slow down.

  2. 02

    Built around the real bottleneck.

    Every engagement starts with the step that is slowing a program down. Often it’s a spreadsheet nobody trusts or a reporting cycle that takes up a whole quarter. Fixing it usually frees up the rest of the strategy.

  3. 03

    Built to last, and to hand over.

    The technology behind this work keeps changing, AI most of all. We release each new capability on its own, then hand it to your team with the training and documentation to run it without us.

The advisory work behind every build See the work

Ports, watersheds, city halls, and boardrooms are where our work happens, and where we’ve spent nearly two decades. The software we build has to work there too.

Examples from CF Labs

A few of the tools we’ve built.

Screens are from the live tools.

Live

Trovance

Our own product, built on the same stack we use for client work. Trovance tracks how AI assistants answer the questions buyers ask, and turns each gap into a ranked list of work with the evidence behind it.

  • Monitors how ChatGPT, Perplexity, Gemini, and Claude answer the questions your buyers ask
  • Turns every gap into a ranked queue of actions, each with the evidence that earned its place
  • Attributes AI-referred visitors to revenue with a first-party, cookieless pixel
trovance.ai
Council Fire · WorkspaceSample data
DashboardVisibilityTraffic
Share of answer39%+6
Answers watched128
Citations won47
AI revenue$18.4k

Waiting on you · 3

Content gapCompetitors rank for “agentic content workflows” — you don’t have a pagePriority 94Accept & draft
Accuracy fixChatGPT gives outdated pricing when buyers ask about your tiersPriority 87Accept & draft
Content gapPerplexity cites a competitor’s guide for “AI visibility audit”Priority 71Accept & draft
In service

Resilient & Ready

A platform for climate resilience planning and implementation. It brings together climate, resilience, and funding data that usually sits in a dozen spreadsheets, ranks projects against the criteria that matter most, and matches them to the funding that fits. The Resilience Authority of Annapolis & Anne Arundel County runs its resilience portfolio on it today.

  • Cross-references climate and resilience data against active funding sources
  • Prioritizes and ranks projects, making the case for what gets funded next
  • Shows impact clearly enough that a funder, a staff planner, and a board member can all follow it
  • Ready to adapt for other jurisdictions
See the Annapolis & Anne Arundel portfolio Read the Annapolis case study
Resilient & Ready Anne Arundel public portfolio: overviewResilient & Ready: portfolio counters, initiative description, and portfolio summary written from live data
In development

Stakeholder Engagement Manager

An AI system that reads an organization’s engagement material and scores it against the stakeholder methodology we’ve refined over nearly two decades, producing a ranked list of the gaps that matter most.

  • Ingests plans, meeting records, and outreach material as they are, with a review step before anything is imported
  • Scores coverage against the stakeholder methodology Council Fire has used in advisory work for nearly two decades
  • Produces a ranked gap analysis a team can act on

Screens will follow once the first client teams are using it.

Intake
Methodology score
Ranked gaps01 · Tribal and community consultation02 · Regulator touchpoints03 · Board reporting cadence

How an engagement runs

Four phases. Working software in your hands within weeks.

Hover or tap a phase.

01

Scope

We find the bottleneck and define the smallest system that removes it.

A few weeks with your team, mostly listening. You leave knowing exactly what to build next.

What we build

Software, dashboards, and AI, shaped to how each program runs.

CF Labs builds what a program needs to succeed, whether that’s software, a dashboard, an AI system, or something new. Most of our work so far falls into four kinds.

  1. 01

    Custom software and platforms

    Client portals, internal systems, and portfolio applications shaped around how your program runs, from intake to board report.

  2. 02

    Dashboards and data infrastructure

    One trusted view of emissions, resilience, or engagement data, with the data pipelines that keep it current.

  3. 03

    AI and agentic systems

    Assistants and automations that read, score, and draft continuously, with a person deciding what ships. We run them in our own products first.

  4. 04

    Tools that ship with the strategy

    Assessments, scorecards, and engagement tooling that turn a recommendation into something a client or a community can pick up and use the same week.

Responsible by design

Building with AI means making real choices: whose data it touches, who it might leave out, what it costs the planet to run. Here’s how we make them.

How we use AI, and what we owe the planet

We build to the Montreal Declaration for Responsible AI.

Ten principles, drafted at Université de Montréal in 2018 with citizens, researchers, and industry at the table. Here is what each one means when CF Labs ships software.

Read the Declaration
  1. 01

    Well-being

    A build has to leave the people who use it, and the communities the program serves, better off. If it only makes a report faster, it is not done.

  2. 02

    Respect for autonomy

    Our systems recommend, your team decides. Nothing nudges a user toward a choice they cannot see the reasoning for.

  3. 03

    Protection of privacy

    Client data is scoped to the people who need it, never used to train models without consent, and deleted on request.

  4. 04

    Solidarity

    We build for everyone touched by the plan, from the signer to the last supplier in the chain.

  5. 05

    Democratic participation

    Engagement tooling is built to bring more voices into the room.

  6. 06

    Equity

    Scoring and prioritization are checked for who they favor before they ship. A model that quietly ranks the wealthiest neighborhoods first is a defect.

  7. 07

    Diversity and inclusion

    Language, access, and local context shape the design from day one.

  8. 08

    Prudence

    We scope the smallest system that solves the bottleneck, test it on real data before it runs unattended, and keep a way to switch it off.

  9. 09

    Responsibility

    A person on your team accepts or rejects every recommendation, and the record shows who did. Automation never absorbs accountability.

  10. 10

    Sustainable development

    Compute has a footprint, and we track it the same way we track cost. It’s the discipline Council Fire asks of a client’s supply chain and has held itself to as a certified B Corp since 2010 and a 1% for the Planet member.

Let’s build it

Bring the plan. We’ll build the system that runs it.

Bring us the strategy that needs infrastructure, or the bottleneck that needs to go. In one scoping call we’ll find the first system worth building and tell you what it takes.