The TLDR

There are 15,000+ wastewater treatment plants in the U.S. Every one of them has operators who look through microscopes to figure out if their biology is healthy. They're squinting at tiny organisms and making critical decisions based on what they see.

We're teaching a computer to see what they see. Our CV model, analyzes microscopy images to help operators understand their treatment process in real time. It's early, it's working, and we need help making it great.

You'd work directly with our lead ML Researcher to train, improve, and scale these models. He drives the vision. You help make it real. Think of it as being the person who turns "what if we tried this?" into "here's what happened when we did."

The Day-to-Day

• Training and iterating on CV models for microscopy image classification and detection.

• Building clean, high-quality training datasets. Annotation, curation, augmentation, all of it.

• Scaling what works from "research prototype" to "production-grade."

• Running experiments, tracking results, and helping build evaluation frameworks so we actually know what's improving.

• Helping move models from the lab into the product.

Who You Are

We're looking for someone curious and brilliant who loves computers, not a specific line on a resume.

• You have some foundation in computer vision or deep learning. Could be school, could be self-taught, could be a side project that got out of hand. We don't care how you got here.

• You're comfortable in Python. Bonus if you've used PyTorch or TensorFlow, but we'll get you there.

• You use AI tools every day to move faster. ChatGPT, Claude, Copilot, whatever. Not as a toy, but as part of how you actually work.

• You're systematic. When you label 500 images, image 500 is just as carefully done as image 1.

• You learn fast. We're going to throw things at you that you haven't done before. That should sound fun to you.

Extra credit: You've worked with scientific or biomedical imaging. You know what activated sludge is. You've built a dataset from scratch and know how painful and important it is.

Who We Are

Nyad is a venture-backed startup modernizing how the world manages wastewater. Thousands of plants still run on clipboards, guesswork, and manual microscopes. With expert operators retiring in a “silver tsunami,” we’re closing the knowledge gap with software. Nyad builds brilliant tools for the guardians of our environment. We love working in one of the least digitized (and most essential) sectors on Earth.

Why This Is Worth It

We're a small team and we're moving fast. Here's what we can offer:

• Your work ships. Not into a backlog. Into a product that real operators use at real plants.

• You'll learn more in a year here than in three years at a big company, because there's nowhere to hide and no one else to do it.

• The problem is real. Wastewater operators protect public health every day and almost nobody is building technology for them. We are.

• You'll be in the room for every decision. This is a team, not a hierarchy.