tomwphillips

>As long as you have queryable data

This is doing a lot of work! Decent model and semantic layer is critical, I think. And of course it has to make sense to the business/domain experts.

Your comment reminds me of this one about Dropbox: https://news.ycombinator.com/item?id=9224

As a data generalist in an business, it makes zero sense for me to spend my time building a it tool like Hex when I can buy one and spend my time actually moving the needle on business outcomes.

Just because you can doesn't mean you should.

I agree. It takes time to develop intuition about how to use it effectively. My skills improved when I adopted a more systematic approach, e.g. using the same prompt across multiple models and examining the results, or slight variations on a prompt with the same model.

I'm surprised at the negative feedback in the comments.

We signed up for Hex a few weeks ago. The PoC was smooth. My colleagues love it.

Agreed.

For ages the data industry has obsessed over tooling when really the biggest impact comes from the organisational/people aspects - achieving consensus on metrics, dimensions, key entities. In short, governance. You can't buy that you like can a SaaS product.

We are now on Cloud Run.

After the last outage we considered moving off to Cloud Run - used it several times before and it is rock solid - but we gave them another chance. Unlikely we are staying after a day of no deployments and then this.

Rotated all our secrets but since deploys are paused our app is down.

I'm unimpressed.

Still down. I think we will move to Cloud Run after this.

You could, but the whole idea of lakehouse is that you can use whatever engine suits your needs. Now you've got to make sure that every engine enforces your access control policies properly. It just seems like a lot of work.

I think this is an elegant design that's superior to the competition, but I think lakehouses are not as generally useful as vendors would like us to believe. The access controls are limited to what's possible on the underlying bucket.

For example I think a lakehouse is a bad choice for standard enterprise BI type analytics - you've got no column or row access controls, and no column masking. I don't see how this could ever be bolted on to the bucket and catalog.

https://www.tomwphillips.co.uk/2026/08/the-benefits-of-data-...

Yeah, me. I've used it every day for years. With a hybrid/remote team it is a fantastic alternative to a physical whiteboard. Whole team uses it.

I don't care for or use any of the AI features.

I hope the price doesn't go up dramatically.

The video demo is so strange. A waitress spoke too fast - just act like a normal person and politely ask them to repeat themselves, instead of reading a transcription on your watch. My goodness.

Definitely not a bubble.

Can you elaborate what you mean by "won't let me fix"? What happens when you try to?

I was in a similar boat with my M1 Pro. I have an M4 Pro for work but rarely notice the difference.

Unfortunately the display in my M1 has failed and a replacement is £500-700. Very frustrating.

*All* my work as a solo consultant/contractor was from former colleagues who needed "trusted pair of hands" to deal with a project, or former colleagues introducing me to new people.

People hire you because they want something done with zero hassle. It is a risk to go with someone you don't know or haven't had someone vouch for.

I have this problem, but I just lock the front door.

A colleague sent me this search directly.

Dupe of: https://news.ycombinator.com/item?id=47570269

Yes.

https://github.com/search?q=%22%E2%9A%A1+Quickly+spin+up+cop...

I'm not sure a staging environment would have caught it.

I often find Claude makes changes that _look_ reasonable, but it's only when I really dig in (e.g. when refactoring) that I realise there's insidious problems.

I can imagine the author making the changes in a staging environment, seeing that it _appears_ to be ok, then blowing up production anyway.

(AI aside, staging is a lie: https://www.tomwphillips.co.uk/2026/01/staging-is-a-wasteful...).

At $dayJob we use (user facing) IDs like this. Select a prefix then add a sufficient number of random alphanumeric characters for your use case.

In October they were talking about it taking 18 months.

https://thenewstack.io/github-will-prioritize-migrating-to-a...

There were reports that they're migrating to Azure. Maybe it isn't going so well. I thought AI was supposed for have solved software engineering...

That's good - I'll reconsider Dagger.

Yes, I agree on your assessment. AI means a higher rate of code changes, so you need more robust and fast CI.

A while back I think I heard you on a podcast describing these pain points. Experienced them myself; sounded like a compelling solution. I remember Dagger docs being all about AI a year or two ago, and frankly it put me off, but that seems to have gone again. Is your focus back to CI?

Also seeing this in the UK right now.

I agree it works well. Although as a long-time TDD practitioner it is mildly frustrating that it has taken LLMs to get more people to realise it works!

Like we don't feed the trolls, we shouldn't the feed agents.

I'm impressed the maintainers responded so cordially. Personally I would have gone straight for the block button.

Anyone know what happened?

Thought it seemed like a great idea but I never tried it. In a startup it seemed like an unnecessary source of risk and in an enterprise too much hassle to convince stakeholders to switch from existing IaC products.

I admired their commitment to open source.

In my last company, we _did_ pay for Google Cloud support and when BigQuery jobs started to fail randomly, causing huge trouble producing critical reports, the response was essentially "we are investigating", "we have identified the issue", and "please wait for it to be fixed". Hardly what I would call support. They couldn't care less.