I’ve been curious for a while: do people actually prefer certain search engines when they can’t see the brand, or is it mostly “I like Google” kind of thing?
Since I couldn’t find a good way to test this, I built Search Bench. It’s like the LLM Arena, but for search engines.
Last year I tried Advent of Code for the first time. I didn’t solve everything, but I liked the format: daily coding puzzles with a story around them. That gave me the idea to build my own version, and to try my hand at making a game.
That’s how Marches & Gnats started.
A couple of years ago I realized social media was grabbing way more of my attention than it deserved. It wasn’t that I was spending hours on it — just a few quick checks of Facebook, Twitter, or Instagram several times a day. The most obvious solution was a website blocker. But those didn’t work for me: they were just browser extensions, and all it took was disabling the extension to get back to scrolling.
In August I’ll join the Antler startup accelerator program in Stockholm, which is aimed at helping people find co-founders and start a company.
As a technical co-founder, it’s important to build an MVP as effectively as possible. This blog post is a reminder to myself about the things I should avoid.
Lately, I’ve been thinking about startup trends. A few years ago, everyone was obsessed with blockchain, then it was chatbots, AI, ICOs, and so on. Every other year there’s a new hype train people rush to jump on.
Many of these “hot” startups end up on Product Hunt. So I decided to parse all featured products submitted since its launch on November 24, 2013, up until June 30, 2019, and see what insights I could get from the data.
I try to unload all information that matters to me from my brain into external storage, because I don’t like to rely on my memory—and honestly, I don’t trust it.
In this blog post, I’ll share my current approach to working with a personal knowledge base.
Over the 8 years I’ve been using git, I’ve tried different GUIs and TUIs, but nothing really stuck with me (I’ll list a few below anyway). Some had a great UI but didn’t cover my whole workflow, while others had tons of commands but the user experience wasn’t that great. So I always ended up back in the terminal, either because it was faster or because the tool I was using didn’t support what I needed.
A couple of years ago I decided to go all in with git CLI and just keep improving my workflow over time.
Recently I shut down one of my projects ListList, and I decided to write a short guide on how to do it properly. There isn’t much practical info about this topic online, so maybe this will help someone.
This guide is about closing a web product, but with some small adjustments it could apply to other kinds of products too.
Recently, I ran into this problem: I’d discover a great source of information, but I didn’t like the way it was delivered. Most of the time, it came as an email newsletter that popped up in my inbox and distracted me. Personally, I’d much rather get it in my RSS reader.
Another example: you might have cron job errors sent to your email inbox, but actually prefer to get them in a Slack channel.