dorkitude

Not exactly. System 2 style reasoning to follow instructions can be expensive and counts as output tokens.

Found the overbuilder!

Same here. Even the page to submit support requests is down.

Cloud console does nothing.

They should host their support services on AWS and vice-versa.

My favorite thing about this article is that it was released the exact same day as Claude 4 Opus.

If it's satire, it's brilliant: because most of the comments I see (here and elsewhere) are clearly written by people who tried agentic coding before Opus 4, and haven't given it a fair shake over the ensuing five days.

IMO the most important engineering skill in 2025 isn't low-level programming, or the craft of debugging, or even having a firm grasp of system architecture. Believe it or not, I truly believe the vibe-first juniors will learn that stuff too, over the course of their careers, just as we did: through necessity (As an aside: if you don't think they'll encounter such necessity, then it's inherently not one any more than the countless other once-honored, fastidious hallmarks of craft that have since been rendered obsolete. And if you don't think they'll learn even upon encountering a true necessity, then you underestimate them.)

No, the most important engineering skill in 2025 is non-attachment: constantly update your priors, and hold your opinions very loosely. Because those opinions could be fully wrong before the essay even gets shared.

This is amazing! Obviously not as good as Perplexity itself, but a great starting point for anyone developing this kind of stuff.

FYI the new terms are previewed here: https://fabric.io/terms-preview

For users of Fabric/Crashlytics/Answers, is there a TLDR somewhere for the new terms? (specifically, the delta between their new terms preview and the old terms)

Great points here. Wish I had thought to pitch the rest of the team first.

For your article, you may be interested in a piece I wrote about Demo Day pitching specifically:

https://keen.io/blog/41079734225/how-to-write-your-demo-day-...

I believe I had spoken with 1 or 2 investors total about the company before that moment.

You're right: practice really seems like the key to getting this stuff down. One can have a clear intuitive understanding of their vision long before they have a clear way to express it in English.

Now that I've finally re-watched it, I can't help but agree. I definitely sounded like a jerk (and probably a fairly stupid one). It's funny how one's defensiveness can be interpreted :(

I appreciate the support, and I'd be flattered if your group were to find this post useful!

Thank you for writing this! We definitely have social game traction, 3.5 years after this pitch. (I come from that industry as well)

It's validating to hear from an industry veteran that our thesis isn't totally off base :)

Our thesis is that storing and analyzing ad hoc data is not a solved problem -- while it may be pretty easy at some scales, it is rather hard at others.

Keen IO takes an API-based approach to the problem, which means:

- developer abstractions are higher level than rolling your own

- scalability is as easy as "my bill went up" (as opposed to "my ALTER statements stopped working!" or "I need to leave mongo and buy a book on distributed system engineering!"

- ability to cover new/emergent use cases is a lot higher than you would get with an off-the-shelf analytics product

The thing about fundraising is, you don't get to choose your audience or their mood. All you can really control is how you react and how you present your business, and you can tactfully drive the conversation where you want it to go.

Now that I've pitched 80-90 additional times, I must say PG's approach in this video is above average, if anything, when it comes to giving the entrepreneur the benefit of the doubt -- so in my opinion it's really not on him.

Wow, what a great write-up! I work at Keen IO and helped design the funnel API -- this is one of the most sophisticated uses of that API we've seen. Makes me proud :')

Question: What's missing from our API? Are there any qualitative or quantitative limitations that got in your way?

Very sorry if I've broken any HN rules. I didn't know those posts weren't HN-appropriate (or even that I had been flagged!).

I don't mind being called out when I'm out of order, because I see it as an opportunity to become a better community citizen. But it's hard to become a better community citizen without some corrective advice.

Would you please point me to the code of conduct (and ideally to the specific rules I've broken), so I can be better in the future?

Great talk from Heavybit!

This post is great.

Couldn't agree more about the four hours per day piece. The best work I did in my career as a dev was the year I worked 20 hour weeks.

I'm glad you had a pleasant experience -- it's too bad it didn't work out.

You may be interested in an upcoming feature we currently have in private beta, which does this:

    your events  ---->  Keen IO  ---->  your S3 bucket on AWS

We have a great, standard format for representing the events in files in S3. If you were using that feature, you could just write an Elastic MapReduce job to perform these sorts of custom calculations (we'll be open-sourcing a bunch of these sorts of scripts). Another option would be to pay one of our partners to write this analysis code for you.

I wonder: would that feature be interesting to you? Would it have perhaps allowed us to win your business?

[Disclosure - I work at Keen IO]

You're right that this is definitely one of our dominant use cases.

It's taken us by surprise to see just how much traction we've among B2B SaaS companies who just wanted to give reports their customers.

Glad the scoped keys are working out for you! If you could add one feature to our product, what would it be?

To see a model for this, look to companies with B2B DNA. Facebook and Twitter are used to taking a "PM knows best" b2c approach, not a "customers first" perspective. Salesforce REST is a good example from larger companies (disclosure - my cofounder was one of it architects), and Stripe's API is a great example from the startup world.

From studying these, I found that the best practices aren't all that surprising: versioning, discoverability, and following HTTP standards.

Systems Programmer / Infrastructure Engineer at Keen IO - https://keen.io (San Francisco only, for now)

Analytics Backend as a Service for modern developers (web + mobile + internet of things)

Our API's usage is doubling every 6-8 weeks. We need help supporting that load.

Keen IO is a fully managed cloud API that lets developers build complex, hyper-custom, large-scale analytics and data science features directly into their web, mobile, or Internet of Things applications.

Our currency is event data, and events are a pretty broad abstraction. Every time someone draws a breath, clicks a link, loads a website, pauses a song, or starts a car, there's an event. Events like these, in large aggregate, are extremely useful -- and therefore valuable :)

It's probably worth mentioning that we're a platform company, not a product company. This means our customers are using the platform to build for an emergent and rapidly growing set of use cases. The same flexibility that makes us future-proof for a given customer also makes us future-proof from a market perspective. So while a product company seeks a certain kind of product-market fit, we allow product-market fit to come to us: our customers build it for themselves.

Things we look for in a new team member:

  • Learning agility 
  • Abstract reasoning 
  • Empathy
  • Potential is more important than credential 
  • Confidence, but tempered by humility & the drive to get even better

Some things we believe in:

  • Mission = The Discovery of Truth.
  • Radical transparency 
  • Design with a capital "D" 
  • Introspection 
  • Distributed innovation 
  • The importance of laughter 
  • Relationships are everything 
  • Play to your strengths; patch your weaknesses with diverse collaborators

Apply: https://angel.co/keen-io/jobs

I love this post!

All too often founders fall into a cycle like this:

Pitch failures begin to beget a habit of fear or panic or other forms of negativity, which of course beget more pitch failures.

I had this happen to me in a bad way once, and it almost cost us our company. But if you find your flow and fundraise when things are going well, it gets a lot easier.

Loved your thoughtful responses!

Love this post. I agree with almost everything you wrote here. One thing I would respond to is the notion that tech industry workers are somehow inherently not creatives or artists. The best programmers and designers I know here are without exception creative, thoughtful people in many other areas of life. The ones who choose SF over the valley may trend even more in that direction.