Automating Climate Research with AI Agents
In this session, you will take a complex climate question and set up a working AI research agent for it. This AI agent will take the multilayered problem, breaks it into sub-queries, searches across dozens of sources in parallel, and delivers a citation-backed briefing you can actually verify. By the end, you'll have a pipeline you can point at your own research domain and run.
Climate professionals spend real hours every week on the same research loops: scanning funding announcements, tracking policy updates across jurisdictions, monitoring literature in a sub-field. The work matters, but it's repetitive, and it's the first thing that slips when deadlines hit. AI agents can take over the searching and synthesizing. You keep the interpreting and deciding.
What you'll see built:
A complete research pipeline using Perplexity's Deep Research mode and NotebookLM. The agent takes a climate research question, decomposes it into 5-10 sub-queries, runs them simultaneously, and synthesizes the results into a structured briefing with traceable sources. We then ground that output in a verified document and use NotebookLM to extract key findings, generate an audio overview, and map relationships between concepts.
What you walk away with:
A replicable workflow for setting up your own research monitoring agent, and the understanding to adapt it to funding tracking, policy monitoring, or literature review in your domain. Plus:
Step-by-step documentation of the full pipeline
Premade prompt templates for research decomposition
Example briefings and output formats
Exclusive discount and access to Terra Studio
The format:
We will build this pipeline live in an hour. I will walk through each decision as it happens, with Q&A every 20 minutes. You'll see what works, what breaks, and how to adapt the workflow to your own research needs.
Go deeper:
The agent workflow techniques in this session come directly from the Terra Studio Intensive curriculum and free courses, where participants build autonomous research and monitoring systems for real climate use cases. The free agentic workflows course is available at studio.terra.do.
Host
Nilesh Bansod Head of Product, COO, Terra.do
Nilesh leads the technical and operations side of Terra.do. He builds the systems that make Terra's programs run.