The Cognavitron — hover for details · click to navigate
Training
Teaching is part of the mission. The tools and methods I use for ecological research are not yet accessible to most of the people who need them — field ecologists, conservation practitioners, tribal environmental staff, agency scientists. Closing that gap is not a side project. It’s infrastructure for the larger vision.
Every course here was designed for people with ecological knowledge but not a machine learning background. The goal is to produce practitioners who can apply these tools to real problems — not just run scripts they don’t understand.
To request a workshop or talk for your organization, get in touch.
Talks
AI 101: What Ecologists Need to Know About Artificial Intelligence
A non-technical introduction to AI for ecologists, conservation scientists, and the people who fund and manage their work. No equations, no code — just a clear-eyed look at what these tools actually do, where they genuinely help, and where they fail. Originally developed for senior leadership audiences; now offered freely to ecology and conservation organizations.
Audience: Managers, program officers, funders, field scientists curious about AI Format: 60–90 minute talk + Q&A Prerequisites: None
In development (converting from PowerPoint) — available on request for non-profit, conservation, and academic audiences.
Workshops
The Honest Ecological ML Workflow
A full-pipeline workshop covering everything most ML courses skip: study design, field data collection, annotation, dataset construction, modeling, evaluation, deployment, and communication. Designed for wildlife practitioners who have ecological data and want to use AI effectively — not just run someone else’s model.
Audience: Wildlife biologists, conservation practitioners, agency staff, NGO researchers Format: 2–3 day intensive (in-person) or split across multiple half-days (remote) Prerequisites: Basic Python familiarity helpful but not required
In development — available for institutional booking.
Scientific Python for Ecologists
Python for field scientists: environment setup, data wrangling with pandas, visualization, and batch processing of ecological data files. Every example uses real ecological data — no generic datasets. A prerequisite companion for the deep learning workshops.
Audience: Field biologists and ecologists new to Python Format: Half-day or full-day workshop Prerequisites: None
Available on request.
Scientific R for Ecologists
R for ecological data analysis: data wrangling, visualization, and statistical modeling with ecologically relevant examples. Developed and workshop-tested at USGS.
Audience: Field biologists and ecologists new to R, or R users wanting to modernize their workflow Format: Half-day or full-day workshop Prerequisites: None
Available on request.
Introduction to Deep Learning with PyTorch
Hands-on introduction to deep learning using PyTorch, taught entirely with ecological examples. Neural network fundamentals, training loops, transfer learning, and evaluation — from first principles to a working classifier on your own data.
Audience: Ecologists and conservation scientists with basic Python skills Format: 1–2 day workshop Prerequisites: Basic Python (Scientific Python workshop or equivalent)
In development.
Introduction to Image Classification with PyTorch
CNNs and transfer learning applied to ecological image data — camera trap images, drone imagery, and spectrograms. Fine-tuning pretrained models, handling class imbalance, and evaluating classifiers on held-out field data.
Audience: Ecologists working with camera traps, drones, or acoustic monitoring Format: 1–2 day workshop Prerequisites: Introduction to Deep Learning, or equivalent
In development.
Deep Learning for Bioacoustics
End-to-end acoustic monitoring pipeline: PAM data collection, spectrogram generation, BirdNET and custom classifier workflows, evaluation, and deployment. Built around real passive acoustic monitoring datasets.
Audience: Ecologists and conservation scientists working with acoustic data Format: 1–2 day workshop Prerequisites: Introduction to Deep Learning, or equivalent
In development.
To discuss bringing a workshop or talk to your organization, contact me.