Industry leaders from utilities, technology, and academia explored how the grid can preserve critical know‑how as veteran workers retire—covering field operations, training, robotics, drones, data practices, and where AI truly adds value.
The theme was simple.
Knowledge walks out the door unless you design processes and tools to capture it—and AI only helps when it’s grounded in your procedures, your data, and your people.
When applied correctly, AI and modern workflows become a force multiplier—helping teams shorten learning curves, improve safety, and turn tacit expertise into shareable, auditable playbooks rather than one‑off fixes.
Watch the full recording, review the slide deck, or read below for key takeaways on preserving and scaling workforce knowledge across grid operations.
First, meet our experts:
Dr. Junia Compostella, ex‑Nissan, Uber, Berkeley Lab.
Erin Westberg, Manager of Cost & Growth @ NextEra Energy Resources
Anto Patrex, Founder & CEO @ CosmicBrain AI
Donald Salmi, Project Manager, Sr (Nuc) @ Entergy
Moderated by @Adam Shaw, CMO — Integ Consulting, Founder — Grid Innovation Hub
Before you get into the summary…
We just opened registrations for the November 14 Grid Technology Coffee Chat — reserve your spot before it fills up.
Share GridInnovationHub.com with your peers, colleagues, and friends.
The workforce knowledge problem
Operators are losing decades of tacit knowledge as experienced craft and control room staff retire. Young hires are often “new to energy/new to craft,” digitally fluent but still building hands‑on skill. Without intentional capture, lessons learned remain in people’s heads and post‑outage reviews get buried.
Stand up a governed “lessons learned” loop that actually feeds the next outage plan
Pair new staff with veterans in structured rotations; protect time for coaching on the deck
Treat field nuance—sounds, smells, troubleshooting sequences—as data to be captured
Why utilities struggle to scale solutions
Good intentions die in PowerPoint. Training isn’t learning without practice, mentorship, and validated procedures. Enterprise AI without plant context misses policy, safety, and site specifics.
Replace static slide decks with interactive 3D, animations, and step‑by‑step job aids
Ground AI assistants in your SOPs, permits, ALARA, lockout/tagout, and site directives
Measure training by demonstrated tasks and error reduction, not course completions
Communities of practice with human‑in‑the‑loop
A living expert network plus AI gets people 90–95% of the way, then SMEs validate the last mile.
Create cross‑site “ask an expert” channels for real‑time Q&A on procedures and edge cases
Merge AI answers with company policies, then require SME verification before use
Add AR/remote assist so experts can guide new techs safely through critical steps
Robots and “physical AI” where it truly helps
Humanoids, mobile robots, and automated tooling can take on hazardous, repetitive, or access‑constrained work—if trained on real processes and safety guardrails.
Start with high‑risk tasks: confined spaces, heights, hot work, internal inspections
Record human task flows, safety interlocks, and “if‑then” decisions as robot skills
Keep humans in control: simulate first, pilot in sandboxes, audit every autonomous action
Drones and advanced monitoring in daily operations
Remote inspections cut exposure and widen coverage—especially for boilers, stacks, blades, and wildfire risk zones—when alerts are validated and trends are tracked.
Use drones and sensors for hard‑to‑reach assets; store imagery with defect labels
Build validated alarms to reduce false positives and fatigue
Mine historical data with governed scripts to surface early failure patterns
Turning tacit expertise into reusable playbooks
Capture “how the best do it,” not just “what happened.”
Shadow veterans with smart capture: head‑mounted video, audio notes, and checklists
Convert recordings into annotated procedures with decision points and cautions
Require citations to source paragraphs and retain version history for every change
Faster training, safer plants: compressing learning curves
Blend baseline knowledge, realistic practice, and hands‑on mentoring.
Baseline: SOPs, animations, and simulations to establish mental models
Practice: guided walk‑throughs, mock‑ups, and AR overlays on real equipment
Field: supervised reps with veterans, then competency sign‑offs
Bottom line: Make knowledge capture part of the job, not an afterthought. Use AI, drones, and robots to extend people—not replace judgment—and operationalize what works so every site benefits.
That’s how utilities preserve critical know‑how, train faster, reduce risk, and make better decisions.
Join Our Next Coffee Chat
We just launched the next Grid Innovation Coffee Chat — happening December 5 at 8 AM PST/ 11 AM EST.
Click below to join fellow grid professionals for a one plus hour conversation on the latest tech shaping the power grid.
👉 Join the Coffee Chat
Thanks for reading, here are some more grid technology resources for you to check out:
👉 Watch the whole webinar here.
👉 Find more grid technology panels, meet-ups, and events here.
👉 Register for our networking coffee chats here.
And if you like our content, check out some of these newsletters we love:
