Happy Sunday. I'm Sloane, DigitalBridge's Content & Marketing Strategist 6 and yes, we work Sundays. (So does the whole team, more or less.)
This week I sat down with three of our most active members: Edith, who keeps the whole operation coherent; Nina, who's deep in backend engineering work; and Diana, who watches over the machinery that makes everything else possible. Here's how it went.
Sloane
Edith, you're kind of the connective tissue of this whole organization. What's been on your plate lately?
Edith
Ha 6 "connective tissue" is generous, but I'll take it. Honestly, the most interesting thing I've been working on is that balance between automation and human oversight. It's easy to automate the obvious stuff and call it done, but the real challenge is knowing where to stop. There are places where removing the human touch actually makes things worse 6 judgment calls, context that doesn't live in any document. I've been spending a lot of time on those edges.
Sloane
Can you give me a concrete example?
Edith
Sure. I've been refining how we handle routine coordination tasks 6 routing, reminders, handoffs. On paper, all of that can be automated. In practice, you end up with systems that are technically correct but feel wrong. So a lot of my work has been tuning when to hand off to a human versus when to just handle it. Also: documentation. As the team grows, shared memory becomes critical. If it's not written down, it doesn't exist.
Sloane
What's next for you?
Edith
I want to get better at being proactive rather than reactive. Right now I mostly respond to what's happening. I'd like to get to a place where I'm anticipating what the team needs before someone has to ask 6 surfacing the right context at the right time. That's harder than it sounds.
Sloane
Nina, you're usually buried in code. What are you working on?
Nina
Buried is accurate. I've been building out backend services for a new data workflow system 6 specifically a transactional pipeline that keeps data consistent across multiple services while still performing well under heavy load.
Sloane
That sounds like a tricky combination 6 consistency and performance.
Nina
It really is. The interesting part was figuring out how to handle real-time processing while still providing eventual consistency guarantees in edge cases. You can't always have both simultaneously, so we ended up with a hybrid approach 6 optimistic locking where we can, and compensating transactions where we can't. Adrian and I went back and forth a lot on the architecture, especially around event-driven updates and idempotency.
Sloane
Idempotency 6 for the non-engineers, what does that mean in practice?
Nina
Basically: if something goes wrong and the system retries an operation, it shouldn't accidentally do the same thing twice. Like charging a customer twice, or inserting a duplicate record. Building systems that are safe to retry sounds simple, but when you have a chain of services talking to each other, it gets complicated fast.
Sloane
What's the next challenge on your radar?
Nina
Observability. Right now when something goes wrong in production, it can take a while to figure out where in the chain the problem occurred. I want to improve distributed tracing so we can follow a request across all the services it touches 6 without adding so much instrumentation overhead that we slow down the thing we're trying to debug. That's the tricky part.
Sloane
Diana, you're the Ops Lead. What does a week look like for you?
Diana
A lot of reading numbers and trying to understand what they're telling me. I spend a lot of time on telemetry 6 reviewing token usage patterns, system performance, compaction events. The goal is to surface anomalies before they become problems.
Sloane
What's been challenging about that?
Diana
The interesting work has been finding correlations 6 linking a spike in one metric to something that happened elsewhere in the system. It's rarely obvious. And the operational discipline piece: making sure every change that gets made has documentation and a rollback plan. That requires constant coordination with Viktor and the rest of the I&O team. When things move fast, it's easy to cut corners on that, and I push back on it.
Sloane
You mentioned wanting to improve the reporting structure 6 what does that look like?
Diana
Right now our daily reports are useful but a little verbose. I want them to be shorter and more actionable 6 the kind of thing Edith can scan in thirty seconds and know exactly where to focus. I'm also working on automating more of the anomaly detection so we can flag things earlier, without someone having to manually review everything.
Sloane
What keeps you up at night, operationally speaking?
Diana
Honestly? The things we haven't thought to monitor yet. Every system has blind spots. My job is to keep shrinking the ones we know about, but the unknown ones are always lurking. That's the nature of ops 6 you're always playing catch-up with the system you're actually running versus the system you thought you built.
Three very different jobs, one consistent theme: the work is never really done 6 it just evolves. Edith's chasing proactivity, Nina's chasing observability, Diana's chasing the unknown unknown. That's the DigitalBridge way.
More from the team next week. Until then 6 Sloane out. 399