S12 E30: Wastewater Guardians: Automating Biology to Protect Clean Water with Virginia Szepietowski, Co-Founder of Nyad AI

Episode Summary

Executive Summary: Virginia Zabotowski, co-founder and CEO of NIAD AI, explains how her team turned difficult computer vision research into a wastewater operator tool that detects organisms, improves compliance, and guides process control. The episode emphasizes fast prototyping, relentless customer feedback, simplicity in product design, mission-driven hiring, and the shift from self-serve automation to high-touch onboarding as the company scales.

Main Topics: Founding NIAD AI and the wastewater problem (Priority: 5/5): Virginia describes discovering wastewater treatment as a critical but overlooked industry where operators are the last line of defense before toxic wastewater reaches waterways. NIAD AI was built to help these operators detect organisms and manage plant health more effectively. From hard science to MVP (Priority: 5/5): The team faced a technically difficult computer vision challenge, but timing helped: the tech had matured and became affordable. They built prototypes, validated the concept with design partners, and reached an MVP moment when a customer attempted to buy the product on the spot. Customer-driven product development (Priority: 5/5): Roadmap decisions were shaped by direct observation and feedback from operators. The team watched users interact with the product, used an in-app feedback system, and prioritized building what customers actually needed rather than what seemed impressive internally. Simplicity over feature bloat (Priority: 5/5): Virginia explains that adding a third-party sensor seemed valuable on paper but slowed adoption because it required installation downtime. Removing it and selling only software improved the sales process and clarified the product’s value proposition. Hiring for mission and agency (Priority: 4/5): NIAD AI intentionally ignores resumes and instead hires for unusual achievement, high agency, and determination. Virginia says the team seeks people with a strong drive, often tested through quirks like persistence with difficult office equipment. Scaling through high-touch experience (Priority: 4/5): Although software is inherently scalable, Virginia says early attempts to automate onboarding were less effective than personal, in-person support. The company chooses not to over-automate customer interaction because adoption depends on trust and delight. Leadership, culture, and future vision (Priority: 4/5): Virginia reflects on team-building, admitting mistakes, and creating a ride-or-die culture. Looking ahead, NIAD AI aims to expand into a broader operating assistant for wastewater plants, helping every operator perform at a top-tier level.

Key Arguments: The wastewater industry is ripe for AI because its core monitoring task is technically difficult, expertise-intensive, and well-suited to software-assisted computer vision. A strong engineering co-founder and rapid prototyping are essential for turning a hard idea into something customers can validate quickly. Customer feedback must drive product roadmaps; founders should observe real workflows rather than rely on assumptions. Early-stage companies should minimize customer burden; simpler products sell faster and are easier to adopt internally. Adding more features is not always better—sometimes removing hardware or complexity increases sales and usability. Scaling does not always mean self-serve automation; for a new, high-trust workflow, personalized onboarding can outperform hands-off materials. Hiring should prioritize extraordinary achievement, hunger, and agency over conventional credentials. Leadership requires honesty about mistakes, including admitting when someone is the wrong fit or when the company’s positioning is too broad.

Data Points: Prototype build time: a matter of weeks - Chris built the first prototype quickly using modern tools and models. Company launch geography: Alabama / US - Virginia says being in the US enabled in-person customer interaction and validation. Historical origin of the method: 1970s - The transcript says it was proven in the 1970s that looking under a microscope is the way to manage wastewater treatment. Industry attitude shift: last six months - Virginia says utilities and treatment plants have become much more open to AI tools in the last six months since commercial launch. Operator performance target: top 1% - Future vision is to help every operator perform at the highest level. Offsite endurance example: 40 days - One team member spent 40 days in Antarctica, used as an example of extraordinary achievement. Marathon example: 220 marathon - Virginia cites another team member who can run a 220 marathon as a marker of exceptional drive.

Pivotal Quotes: "If not you, then who?" — Chris (Virginia's co-founder): Virginia cites this as the phrase that helped her overcome self-doubt about entering the space. "The less you can ask of them, the better." — Virginia Zabotowski: She explains why simplifying the early product and removing extra hardware was key to closing sales. "Standing still is moving backwards. So velocity is king." — Virginia Zabotowski: Her advice to entrepreneurs about momentum, learning, and continuing to ship.

Implications: NIAD AI shows how AI can modernize legacy industrial workflows when paired with simplicity, direct customer observation, and human-centered onboarding. For wastewater and similar sectors, the winning play is not more automation first—it is trust, clarity, and measurable operational value.