Andrew Askins

We've made some big improvements to the MetaMonster crawling and onboarding experiences over the last month! You no longer have to wait for a crawl to finish to start queuing up your first workflows. This makes the overall experience of getting started with a new project WAY faster. Next up:

- More guidance and education on your first run - Easily test workflows on a small subset of pages so you don't burn through your credits - Add workflows to a queue so you don't have to wait on one to finish to queue up the next

Three years ago I sold my agency to one of our long time clients and went to work for them. I made it 11 months before I got fired. I wrote about the experience, how it felt, and what I took away from it: www.andrewaskins.com/how-to-get-...

I tried to write about this as soon as it happened. But everything was too fresh, and the drafts I wrote weren't coming across the way I wanted them to. So I've let it sit and percolate in the back of my head for the last 2.5 years, and last week I sat down to take another crack at it.

Something we've found helpful when testing prompts for MetaMonster: have the LLM output it's "rationale."

Maybe it's weighting a certain word heavily, or associating it with something you didn't consider. (at least I think that's what's happening) The new UI makes it easy to generate this at the same time as the output from your prompt inline, and then review them at scale.

Something we've found helpful when testing prompts for MetaMonster: have the LLM output it's "rationale."

Rolled out handy little feature in MetaMonster this week to let you export all of your page content as markdown.

You can use this to help with a site migration, or load all of your website content into another AI tool to run prompts against it. I like to use this personally to add website content into @nbaschez.bsky.social's tool lex.page so I can provide more context to the AI when I'm working on content.

I think I may have taken the "bootstrappers need to charge more" advice a little too seriously and overpriced the MetaMonster beta... 😬 Don't want to overreact to small sample sizes just yet, but seriously considering cutting the price by 50% (add increasing what's included in each plan)

Really appreciate the real talk from @leenyburger.bsky.social in the latest episode of Tiny Seed Tales www.startupsfortherestofus.com/episodes/epi...

I've been feeling the weight of hitting month 6 on MetaMonster lately We only have 1 customer and while I like the product, it's not where I hoped it would be at this point Combine that with a slow month of progress and just *gestures at the world* and feeling my emotional runway dropping a bit

We just released directed crawls to alpha users! 📋 Now you can choose to crawl your entire site or crawl a list of pages you specify This is a great way to narrow suggestions only to the pages you care about, or break a large site into batches

To kick off a directed crawl you choose "crawl selected pages" from the sites page and then upload a CSV. Our monsters will try to identify the column with URLs, but you can also change it easily if you need to.

Hot take: I don't hate Jira They've put a lot of work into updating it, and it's a powerful tool that can be really effective in certain circumstances

If you can't build a good remote culture that's on you, not your employees. Honestly the push back on remote work makes me want to build a team again. It's going to be so easy to compete for great people and give them a better environment than 90% of what's out there.

A LOT of leaders taking the wrong lesson away from the last few years. "Man, this remote work thing is hard, we must need tighter control over our employees." The correct lesson? Your processes suck, your culture sucks, and money is only motivating to a point.

I dig this essay from Edmar Ferreira about the differences between building in AI versus traditional software startups. every.to/source-code...

And his breakdown of AI startups into: - Deep AI startups - building foundational models or hardware - Applied AI startups - which leverage existing models and APIs

I dig this essay from Edmar Ferreira about the differences between building in AI versus traditional software startups. every.to/source-code...

In general, I agree with his argument that AI startups have a lot more feasibility risk than traditional SaaS startups. The technology is still new enough that there are a lot of questions about what it can do (and a lot of people making bullshit claims).