Hi Charlie,

Thanks for the note, and for building Atlas into something many of us rely on daily.

I know this question is going out to a huge developer base, so instead of a feature wishlist,

I wanted to share one idea that’s been on my mind as a builder heading into 2026.

MongoDB is incredibly powerful, but for many developers (especially those growing into system-level thinking), the hardest part isn’t writing queries; it’s knowing whether their data model and usage patterns are “right” before things break.

My suggestion for 2026:
Make Atlas more opinionated as a “thinking partner” rather than just an observability platform.

Concretely, imagine Atlas offering:

• Proactive, human-readable insights like:
  “This schema works now, but will likely cause performance or flexibility issues at scale because…”

• Context-aware suggestions that connect data modelling, query patterns, and real-world workload behaviour, not just metrics.

• A way to simulate or preview “future pain” (growth, traffic spikes, feature expansion) before it happens, especially for early-stage teams.

With AI tooling improving fast, developers aren’t just asking “Is my query slow?”
They’re asking, “Am I building this system in a way I won’t regret in a year?”

MongoDB is uniquely positioned to answer that question because it sits at the intersection of data, application logic, and scale.

If Atlas could help developers learn and reason, not only diagnose, it would quietly become one of the most valuable educational platforms in modern software development, especially as more AI-generated code enters production.

Thanks for listening, and for continuing to build for people who care deeply about doing things the right way.

Wishing the team a strong 2026,
Zirwa