In tech, getting feedback that is both compassionate and actionable is still surprisingly rare. Too often, it’s either vague encouragement or blunt criticism — neither of which helps people grow. But what if we made high-quality feedback the default operating system of our teams?
AI has forced us to get serious about two things: observability and evaluation. Observability means continuously monitoring real-time behavior, surfacing insights into performance and health. Evaluation is the structured process of analyzing that data against specific goals. Both are essential. Without observability, you miss signals. Without evaluation, you miss meaning.
Humans need both, too. Observability for people looks like regular check-ins, listening, and noticing day-to-day patterns — how someone is showing up, where they’re struggling, where they’re thriving. Evaluation is the structured reflection that turns those signals into growth. When leaders neglect one or the other, growth stalls.
If AI needs observability and evaluation to function, shouldn’t we give humans the same care?
Only “observing” without clear feedback feels like being watched without support. Only “evaluating” without ongoing care feels like being judged without context. The magic is in combining both — continuous attentiveness and structured, compassionate guidance.
If we want AI and humans to thrive together, we need to lead with smarter feedback cultures: embed care and clarity in every piece of feedback, make it continuous, anchor in outcomes, and treat feedback as investment. Observability keeps AI safe and reliable. Feedback keeps people resilient and inspired.
The future of work won’t just be defined by AI models improving. It will be defined by humans and AI both having the feedback loops they need to thrive.
Ian Alexander
VP of Design — writing on leadership, AI product strategy, and building teams that ship.