Climate and public technology
Fire, deforestation, monitoring — systems that governments, NGOs, and communities actually operate, rather than pilots that end with the press release.
Antonio Leblanc — Rio de Janeiro
I inherited a project as an intern. Today, Pantera is an AI wildfire platform protecting 40 million hectares with 190 cameras.
Intern to CTO in ten months. I lead Pantera end to end at 1.5°C — 95% detection accuracy, including the Pantanal. Policy, community engagement, and ethics matter just as much as the model — that's where I want to keep spending my energy.
- Pantera
- Inherited from a teammate who left to do a PhD at NASA. I lead it end to end today: architecture, detection, risk, response, impact.
- 1.5°C
- Co-founder and CTO since 2020, Rio de Janeiro.
- Scale
- 190 cameras, 40M+ hectares, 95% detection accuracy — including the Pantanal.
- Agents
- Eight AI agents run sales, support, and marketing on Hermes, open-source, on our own servers. One alone has opened 400+ issues, about 25% closing without real review, mine included.
- ForeFire
- Maintainer since 2022 of CNRS's open-source wildfire simulator, and co-author on its JOSS paper — reviewing and merging PRs, cutting releases, Dockerization, CI/CD, documentation.
- Now
- Open to a technical second opinion for early-stage climate teams, or to hear what you're building in wildfire, deforestation, or agentic systems.
Where I focus
Agentic systems outside of AI companies
The hard part isn't the model — it's putting it inside an operation where being wrong is expensive.