# Rachel Okafor, Senior Engineering Manager at Crescent Health (62 people) — read of DeltaDB, June 12 2026

> 11 years in software, last 4 as EM. Currently managing three squads. I take Caltrain in from Redwood City and I listen to one podcast per leg of the commute. My 8-year-old thinks I work at "the computer place."

## How I got here

Someone dropped this in the #tools channel of Rands Leadership Slack with the message "huh, interesting." No commentary. Just the link. That's the exact kind of share that gets me to click because it means the person couldn't decide how they felt about it either. I had 6 minutes on the train so I opened it.

## What I clicked first

The hero pulled me in fast. "See Every Change, Replay Any Moment" is actually a pretty clean promise. I've lived through the exact situation this is describing: you come back to a broken test at 4pm and you have no idea what you did at 2pm that caused it. The spec table reads almost like a features list from my own internal wish list. "Team Context: New engineer watches your session instead of asking questions. Onboard in hours, not days." I stopped on that one. I've spent probably a combined 40 hours in the last year doing exactly the thing that sentence is trying to replace.

## Where I paused

About halfway down the page there's a section called "How honest is this idea, really?" and then it says: "Honest disclosure: we don't have live customers on this idea yet." I had to read that twice. So this is not a product I can sign up for. This is a product IDEA that someone is selling me a dossier about, for $5. And for $99 I can buy starter code and brand assets and "outreach packs." I am not the customer who is supposed to use DeltaDB. I am apparently the customer who is supposed to BUILD DeltaDB. The page did not make that clear until I was halfway through it.

## What I distrusted

The scoring system. The page says the landing page quality is "2/10" and the financial upside is "1/10" and then immediately below that it's trying to sell me this same idea for $5 to $199. There's something a little dizzying about reading a page that tells me the page I am reading is bad, and also asking me to pay for more information about the thing on the page I am reading. I actually respect the transparency, but it also makes me feel like I wandered into someone's lab notebook rather than a product I can use or buy. The line "We shipped the strategy package; you ship the customer conversations" is doing a lot of work. That is a polite way of saying "we thought of this, you figure out if it's real."

Also: the Fermi estimates. "$-29,500 Year-1 take-home" and "1 in 8 Meaningful-success odds." Those numbers feel precise in a way that hides how made-up they probably are. I don't know what model produced a Fermi estimate of exactly negative twenty-nine thousand five hundred dollars but I'd want to see the inputs before I let it influence a decision.

## What would convince me

If the DeltaDB IDEA is what I should be evaluating, then I want to see one or two recorded conversations with developers who were shown a prototype and asked to describe their current workaround. Not quotes, not paraphrased responses. Raw transcript or video, 5 minutes. That would tell me if the pain point is real enough for someone to change their workflow. The spec table describes what I WANT to be true. Conversation clips tell me what developers actually say when this problem is in front of them.

If there's a working demo of the session replay concept (even rough, even one use case), that beats every word on the page.

## What I'd ask in an email reply

1. The page says "Zero overhead" under Continuous Capture. What does that mean in practice for a 300,000-line Python monorepo running pytest? Where does the session data live and who controls it?

2. You're calling this Git-Native and say it "exports to standard tools." Which tools specifically, and what does an export actually look like? A git patch? A JSON blob? A video file?

3. Is there a version of this I can actually run today, even in alpha, or is the $5 dossier the only thing that exists right now?

## Verdict: on-the-fence

The underlying tool idea is the most genuinely interesting developer productivity concept I've seen in a while, and the spec table describes a real pain I deal with regularly. But I came here thinking I could sign up for something, and I can't. The page is selling the idea of building DeltaDB, not DeltaDB itself. Those are very different products with very different buyers, and the page conflates them enough that I almost left before I figured it out.

---
*Memo by skeptic persona, generated 2026-06-12. Studio breaks own self-grading loop.*