The fact that we’re all talking about it, and not at all surprised, is a great example we can take when making the case for more 9’s of reliability.

* well, very technical power users.

Experienced Go architect (15 years; Zalando, StreamElements). I untangle fragile microservice architectures and make them boring again.

I’ve modernized fragmented microservices for hyper-growth companies, migrated Riak → Postgres under hard deadlines, and turned chaotic legacy systems into stable, scalable platforms.

Recent impact

Overhauled business-critical messaging systems for high-volume, low-latency workloads. Simplified/retired 11+ microservices, reduced latency, and improved observability — enabling a hyper-growth partner launch ahead of competitors.

Led discovery, design, and delivery of an auditable, multi-source, high-throughput inventory tracker — giving an e-commerce client immediate stability and a scalable foundation for years of growth.

Migrated an auth database from Riak → Postgres under a hard deadline (World Cup traffic). Zero downtime and smooth scaling through peak load.

Ideal clients

Berlin/EU scale-ups with distributed systems complexity — fragile microservices, data-intensive platforms, or legacy architecture that's become a liability.

€100+/hr retainer. Fixed-bid projects also considered. Available now for 3–6 months.

The usual stack

Go, Postgres, Kafka/NATS, Kubernetes, AWS/GCP

Get in touch

Email: barry@barryhennessy.com

Portfolio/CV: barryhennessy.com

- someone finally reading the T&Cs

- legal drafting the T&Cs as broadly as possible

- the actual systems running at the time matching what’s in the T&Cs when legal last checked in

Maybe this is a point to make to the Persona CEO. If he wants to avoid a public issue like this then maybe some engineering effort and investment in this direction would be in his best interest.

But I’ve seen a lot of similar claims - just open LinkedIn for a second - and I always come back to the same questions:

- What value has been delivered?

- How much did you spend?*

- How long did it take _all told_?

I know you made a context management db. But if your argument is that AI is the future like this then that seems a somewhat self-referential proof.

What value has been delivered/products built outside of tooling to build products?

I’m aware you probably can’t be 100% open here - IP and all - but I feel it would go a long way to reinforcing your arguments the more concrete you can be.

* points for being up front about the 1000 per engineer minimum. But there’s still the human cost and actual token cost here

Check

"calendar systems suck. All of them."

Double check

"I’m trying to break off of big tech as much as I can"

I wish I could check this more ...

I've had similar needs/desires/gripes with my calendars and the terrible state of calendaring apps for a while. So thank you for scratching your own itch and sharing it with us.

I'm curious when you say that "[CalDAV] is an area begging for disruption". Can you enlighten us as to what your wishlist could be for (a) better protocol/systems/ecosystem might be? (a rant about your pain points might work too).

Thanks again!

I'm curious as to what has been built (by yourselves or others) in the 4 (?) years since its release. Have the experiences held up to the initial goals and designs?

I'm super curious about realtime multiplayer solutions (and I don't think I'm alone). But I find a great lack of info on what running this kind of app would cost. I come from the old-school hold-no-state, request->response->gtfo mentality, and I always have the _feeling_ that it'd be expensive to scale. Not just holding the websocket open, but how much effort do you expend parsing the WAL? How chatty is that kind of persistence mechanism? What other 'gotchas' are there etc.

I'd love nothing more than to dispel that vague feeling with data. I know it wouldn't come close to a full performance analysis, but throwing a few datapoints on a chart would help get a ballpark idea and tune my hype->action conversion.

I could nail down a pile of questions, but I'm sure you know better than I how to measure your own systems. But roughly I'm wondering: - how many users did you have? - how much traffic did you get? - how much would/did it cost to host on supabase? - how much resources did the database/realtime components consume?

Congrats again on the launch and have a nice weekend.

It's complicated enough that I could easily miss a foot-gun, and clearly a sales tool: there's plenty of wiggle room built in to how prices are calculated in the end.

Have you ever compared your actuals to the estimates you started with? Were they accurate? Miles off? If not how then did you budget for cloud costs? Or did you skip that and just see what happened?

Same question goes for the other clouds/providers etc.

Thanks for your time!

Comments URL: https://news.ycombinator.com/item?id=31396466

Points: 1

# Comments: 0

But I always come to the same question with services that provide auth and user management: You pay a lot of money for someone _else_ to own critical information about your customers. What happens if you want to move away and use a different/your own/your customers own service?

Your customer data (at least login) lives in WorkOS' database. How do you get it out? How much does that cost? Are there contractual guarantees around that?

The same goes for your customers integration points. If the customer has to do any setup to integrate WorkOs for your app then moving away would involve them making changes. Not necessarily an easy thing to manage.

Not to be negative: I'd be happy to hear that WorkOS have great processes and guarantees around this.

How you manage decisions in your teams? Do you have heavy process, lightweight, none? What works, what _doesn't_ and in what situations?

My aim here is to flush out ideas and generate some discussion. Hopefully to figure out some better ways of managing this critical piece of our work.

Thanks for your time!

Comments URL: https://news.ycombinator.com/item?id=31303198

Points: 1

# Comments: 1

* Tracking (and auditing) who is receiving what messages.

* Let customers manage their preferred contact method.

* Manage templates and UI centrally (e.g. email templates).

* Track GDPR and consent.

* Block if consent has been revoked.

For example: a basic web app with a keycloak server to handle auth (registration/password reset etc.). Keycloak has a simple smtp integration and hooks to customise templates. But, naturally enough, the web app needs to send emails too.

Is there a more scaleable way to manage consistency and tracking than 'just do it manually'?

Comments URL: https://news.ycombinator.com/item?id=31237765

Points: 3

# Comments: 0

How have you solved this in the past (or present)? Is there a service/project that handles this smoothly?

My gut tells me that this problem has been solved many, many times before, and that someone may have written a service for it, but I can't seem to find anything.

The obvious choice seems to be the open formats: CalDAV, iCalendar etc. But I read nothing but bad things about integration with the likes of Google Calendar. From data loss to 12-24 hour delays in syncing.

So that pushes me to API integration. But integrating with every calendar API out there seems like a never-ending job.

Any insights and help are welcome.

GCal data loss: https://evertpot.com/google-carddav-issues/

GCal delays: https://help.cheqroom.com/en/articles/1502123-why-is-my-google-calendar-sync-slow-or-not-updating-at-all

Comments URL: https://news.ycombinator.com/item?id=31190612

Points: 1

# Comments: 0

The longer you can scale the product without having to scale the application, the better!

These sound like the kind of hidden costs that could turn sqlite's simplicity quite complicated if you don't see them coming.

I.e. it's leaning more toward the moderate, but reliable writes, and heavy read use cases?

Please let me know ~if I'm missing anything~ what use cases I'm missing.

I guess the take away here is that this underscores that sqlite isn't for the 'large number of writers' scenario.

p.s. > I didn't want to be the guy who comes in and says "you are doing it wrong" month 1 Very wise

Out of interest, were you running on bare metal/cloud? And what kind of CPU was behind those 24M face compares per second?

I've never (deliberately) considered committing a DB to git. Although there was that one time when I was straight out of college...

Pro tip: surprising your colleagues in the morning with a 40 minute wait to pull master (because you committed a ???GB db) is a good way to feel like a right eegit.