Recent submissions

Comment · 1 month ago
Comment · 2 months ago
SlimeBallBench was not me, but it is cool! and at least partially what inspired me to build my own.

Mine is less fun, but hopefully more useful https://erikwiffin.github.io/memory-reasoning-eval/

Selling to labs is more than I'm looking for. I'm aiming for a couple hundred dollars so I don't have to finance a Fable vs Sol run out of my own pocket. It would be cool to have my benchmark be one of the ones referenced in a model card!

Comment · 3 months ago
I've developed a benchmark that I think should be resistant to saturation, is easily verifiable, and anecdotally correlates with desirable behavior (ability to not get confused while generating text with state).

I think it's interesting, I think other people would find it useful, but I don't want to spend a bunch of money running it against all the frontier models.

What's the best way to reach out to labs like yours to collaborate on something like that? Are there any labs that are more open to submissions from internet randos?

Comment · 3 months ago
I actually started with it as a PWA, but ended up rewriting it in react-native. I wanted a couple features (heart rate monitor, lock screen integration) that afaik don't work with PWAs.
Comment · 3 months ago
Ugh, that happened to me too. But instead of researching other apps and waiting for them to decide to go premium too, I built my own workout app. Been using it for a couple months now and it's working great, but the hassle of publishing it to the app store means that it will only ever be a solution for me and I can't share it with anyone.
Comment · 4 months ago
As a diver, I'd love to use this and show my friends, but it's a bit hard to get to actual results for a location. Particularly for sharing, if the URL changed to something like `/diving/location/{lat},{lng}`, it would be helpful so I can drop a link in a group chat for "here's the conditions we'll be seeing later".
Comment · 9 months ago

  Location: USA
  Remote: Yes
  Willing to relocate: No
  Technologies: Python, Typescript
  Résumé/CV: https://erik.wiffin.com/
  Email: [email protected]
Fractional Tech Lead with 17 years of tech industry experience. I work with early stage and bootstrapped companies to help them establish a solid technical foundation. Whether that's building your MVP, your first engineering hire, or setting up an SDLC that gets out of the way so your team can focus on what they do best.
Comment · 3 years ago
When I first moved into management, I wrote an article about how my direct reports could get more value out of our 1:1s. I'd probably write something different now, but the broad strokes remain true. https://erik.wiffin.com/posts/how-to-get-the-most-out-of-you...

1. Don't make it a status update. It's your time, and that's probably not the best use of it. 2. Spend some time figuring out what you want to get out of our 1:1s. If you don't know, tell me that, and I'm happy to work with you to figure it out. 3. End the conversation with next steps/action items. It's easy for me to say something that you interpret as a promise even if I didn't mean it that way and calling out action items explicitly is a good way to keep that from exploding into a huge problem.

Comment · 4 years ago
This is asking a lot from you, but in your situation, I think your goal should be to build up enough of a relationship with your manager that you feel comfortable telling them that, or realize that that's never going to happen at your current company and find a new job.

I have reports who have told me that they're happy doing what they're doing. I don't see it as a negative at all - they're employees who are quietly getting their job done and I don't have to worry about them leaving for career growth or Peter Principle-ing them into incompetence. It's important that we continue to have 1:1s though. People aren't static, they can change their mind, and I also want to make sure I keep an ear out for little things that are annoying them, and do what I can to make them go away.

Comment · 4 years ago
Ironically, the feature that keeps me from using Goodreads is that I want it to be less social. Do you have the option for private bookshelves?
Comment · 5 years ago
If Bob had a plan for doing the work that didn't involve Alice's library, there's a good chance he would have just jumped straight in to implementation without talking to anyone, and maybe the library wouldn't have come up until code review (if at all). By identifying the complexity mismatch ahead of time, they saved 3 points worth of effort. This doesn't affect sequencing at all, but still seems valuable to me.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton