tomgs

Yep, would be very cool and it was a great live call

I joined the chat as well, but if Hendrik is here - we'd love to have you on the channel (https://youtube.com/@WithMultiplesAI). This will make for one hell of an episode, I think.

Hey! Getting a 502

The level of "fake it till you make it" in this website is legendary.

Great idea!

You have not social share preview image on the homepage: https://www.opengraph.xyz/url/https%3A%2F%2Fwww.cyberdesk.io...

Concrete use cases where 50 percent is actually a thing?

Neat! What’s the use case exactly? Kinda hard to figure from skimming

Sahil did a thing: https://x.com/sahilypatel/status/1935745481898180991

Looks about right. see 6

Correct.

They weren't there, and then they were there.

I can say, at least from looking at it from up close, that it doesn't seem like paid ads did the trick here. It was a community play, all along.

Don't have proof, but the discord community and the WhatsApp groups tell a decent story.

See my other comment in this thread for more comments.

I helped out in their recent hackathon - https://base4good.com/ - mentoring folks on the app, and I also admin one of the user groups.

I am not a paid member of the team, just an admirer who wanted to get closer to the action. This felt right to me from the first moment, and I'm happy I had a small part in the journey.

I met Maor (the founder of base44) and team and had beers with them. Good people.

--

Let me clarify a few things:

1. I don't know exactly how much vibe coding went into building base44 itself. I can attest that Maor's rate of releasing features was absolutely insane - I'm talking major updates every 1-2 days. I assume he's good with Cursor and the like. He's also very, very decisive on what to build and what not to build. Aggressive, even, I would say.

2. Maor had, for the majority of the life of this, no team. The employees joined way after base had customers. Most of what Base is was built by Maor, with 1-2 close friends helping cut out everything that wasn't relevant or wasn't great (so I'm told).

3. It's a different take on lovable/bolt etc. No one argues this.

4. Maor opted to include the db within the platform, rather than enable persistence externally. This really made the output great, and made fixing cross-application things very easy.

5. To me, base44 is PHP. It's a bit ugly, but it works, easy to explain to people, and once you get a hang of it it's a great hammer. It's not going to win the space race anytime soon, but it'll build you a house.

6. Base has resolve with AI functionality, which is far superior to anything I've seen outside of an IDE. It just works.

--

To folks trying to win the AI race by building exceptional technology on the bleeding edge, good on you. I don't think Base is that.

I think Maor symbolizes something different: we're in the fast-grab era.

Big cos are not able to build killer AI apps at the rate they're expected to, which means they're circling around looking for what they can snatch with money/equity.

My take?

Build AI things that just work for a specific use case. Release them fast. Make people fall in love with them because they "do AI" for the use case.

Some bigco is flying close by, trying to build it but failing. Be there for the purchase.

Hey folks,

I made this little app that goes through the Hacker News Algolia API (https://hn.algolia.com) and gets the least upvoted posts in Show HN every day.

Basically a way to give second wind to people for their projects.

I made it as a test run for Base44 (https://base44.com) - no commercial affiliation, just a fan - and it performed well.

Would be happy to get some feedback!

Location: Israel

Remote: Yes / Hybrid

Technologies/Skills: GTM for very technical companies, technical product marketing, making sure your tech wizardry is actually being used by people.

Résumé/CV: https://granot.io (goes to LI)

Email: LinkedIn should be fine. If you search for it my email is also there.

I'm an engineer turned marketer, former devrel and VP Marketing.

I've spent a while in consulting, and considering (as in: still contemplating) joining a build again. Looking to join (as a co-founder or early founding team) companies that:

1. are working on something that makes a real impact in the physical world

2. are working on something that is highly technical (not traditional B2B SaaS or B2C)

I'm very good at making early-stage companies look like Series B companies, and making sure they say the right things when talking to the people they sell stuff to.

I'm reading it as "your brand will only hold you for so long if you don't keep at it"

No, he doesn't, item 6 on short-term moats:

6. Reputation / Brand: Building a strong reputation often directly boosts sales, and AI is likely to make the brand-building process easier in surprising ways. Having a brand with a rich history can also be an advantage, given you consistently keep working on it and maintain its value over time.

Location: Israel

Remote: Yes / Hybrid

Technologies/Skills: GTM for very technical companies, technical product marketing, making sure your tech wizardry is actually being used by people

Résumé/CV: https://granot.io (goes to LI)

Email: LinkedIn should be fine. If you search for it my email is also there.

I'm an engineer turned marketer, former devrel and VP Marketing.

I've spent a while in consulting, and considering (as in: still contemplating) joining a build again. Looking to join (as a co-founder or early founding team) companies that:

1. are working on something that that makes a real impact in the physical world

2. are working on something that is highly technical (not traditional B2B SaaS or B2C)

I'm very good at making early-stage companies look like Series B companies, and making sure they say the right things when talking to the people they sell stuff to.

Small professional rabbithole:

So I do GTM and work a lot with marketing websites for companies with long sales processes.

The company mentioned there, Supercreator, funnily has a CRM - which is not just some funky AI chatbot thingie, but a proper enterprise thing that people use when doing sales.

It looks like they're treating creators and the "agencies" (whatever that means) as what we would potentially call "SMBs", and sell this CRM thing to them to manage their "customers", which I assume are the fans or subscribers or whatever on the OF side of things.

This is insanely interesting to me. Look at the website - you have a "request a demo" section (which is super enterprise B2B), look at the menu, it's like an enterprise SaaS website.

What the actual fuck is going on here lol

Very cool, thanks for sharing! I have writing work on the side for engineers, if you know any. Great way to get your writing skills going while getting paid to play with tech.

I actually started at DevRel and quickly understood that if you swap writing blog posts for writing whitepapers and website copy, you can easily make significantly more money.

Also, better and more understood KPIs - if I made this really good solution diagram and slapped it on a one-pager that I templated, and then the AE sent it to a customer and it made a sale move 1.2x faster, that's tangible results.

And the numbers don't even have to be insane; I remember building a dashboard as a head of DevRel that was made entirely out of vanity metrics. Today, I don't even make dashboards anymore - the results of my sales collateral are simply evident by the pipeline moving forward.

Also, I'm sorry to say, but so many DevRels are primadonnas - they won't go on sales calls, but they'll go to conferences. If they don't go to a conference every month, then they're not "getting what they came here for". Demand generation is beneath them, since "I don't look at leads, I'm building a community". Bleh.

Much better to work with people - as cutthroat as they may be - who push me to create better content that drives results.

Also led me to creating my first productized service - https://syntaxcinema.dev - and that's been going very well for me recently.

Love this - went ahead and implemented the comments with a few extra niceties.

Thanks brother!

Can one ever really be long on new business? :)

As the greatest singer in our generation said so succinctly in one of her latest hits:

Yeah, my receipts be lookin' like phone numbers If it ain't money, then wrong number

It's kind of like that for me, if phone numbers were five digits!

Oh, I should have 100% said this instead.

Yeah, yeah, scrap what I said - let's go with Pete.

As an aside, what I normally do is reply to each form submission with a personal video and an overview (albeit a short one) of what I think the project should look like given their demands. I then drop a price and get on a call, if they want to.

Many do! And the ones who don’t still get a taste of how I think like and perhaps want to talk some more about other things - this is a side-business productized service; I mainly contract with technical startup companies to do their GTM.

It’s actually looking like it’d be a pretty decent lead funnel for that too!

Man, going out on my own was a great fucking decision.

Oh, half the fun is the sales calls. It’s very much something I can delegate off to a person on my team, but this is relatively new and so the volume is not that large yet. I also really enjoy them!

I treat these more as discovery sessions than anything else. It’s also how I’ve landed on the exact pricing points I have - talking to people, especially when early, is a great experience.

There’s also something to be said about showing prices only in calls - people hate it here (or in general), but there is value to showing the price at the end of the call and not at the start. You get to show the thing to the person, get them excited, and then the price point looks a little different.

The trick is tailoring the sales call to the person - if I can convince them that they should keep me around since I provide a good service, they might pay the price.

Also, there are always discounts as the guy above mentioned;)

Oh definitely low four figures. Can’t imagine doing it for five figures (maybe I should imagine harder) and for less than that it’s just not an interesting pursuit financially for me (given the amount of work required and how much I think my time is worth).

Right here in the thread so everyone knows that I offer discounts, clearly! :)

I actually do that as a service for companies:

https://syntaxcinema.dev

I think that product tutorials are somewhat of a black art. On the one hand you have:

1. Keeping the flow moving and the video fast-paced and interesting

2. Adding aftereffects and other visual niceties

3. Pointing out the relevant bits with zooms, highlights, etc...

But on a deeper level, you also have questions of:

1. Am I using the right sample app to demonstrate my use case?

2. Is the feature I'm using bulletproof? Do I need to change something in the DOM of the application since that feature is not 100%? Do I need to not show a piece since it's irrelevant? Do I need to speed through or flip over from things while they're running / fetching / compiling / generating etc...?

3. And, maybe most importantly, what is the message I intended to deliver? Is that a product overview? A documentation-oriented video? A demo for a conference or a customer? Who's my audience? Am I speaking to them?

I've been doing videos for a while, and I found that the second part of the problem is actually not as easy as one would assume.

I applaud great YouTubers for that - they cracked how to do walkthroughs of products that are not only technically interesting, but also visually pleasing.

I'm a bit of a video nerd, I guess. I started out way back when doing these little nuggets of absolute terribleness (oh my god the thumbnail) https://www.youtube.com/playlist?list=PLlM7w0mARnn4ytxM6s-0b...

And happy to say I improved a little bit from then :)

(that website's pretty new, comments more than welcome)

Following one of the comments in this thread, I reviewed two other products in this space - https://www.staticcms.org/ and https://decapcms.org/.

It looks like the webpages are almost a direct copy of one another, one in dark mode and one in light mode, one with a community strip and professional services and one without

I'm a technical product marketer, and I find this type of landing page copying amusing to no end.

A quick one :)

While we're on the topic - how does traceloop factor into all of this? What's the connection between the two? I assume the former is the LLM observability platform (Datadog for LLM?) and the latter is your own auto-instrumentation thingie to supplement it?

Cool! Two questions:

1. Where do you see this observability for LLM thing going? What's the end game? Is it like in traditional observability where all formats eventually will converge to one format (which OpenTelemetry is trying to be)? I feel it might be a little bit early to tell, tho

2. I noticed you do auto-detection of the framework used, like LLamaIndex et al. Except for annotations, is there a deeper connection to the LLM framework used? This is auto-instrumentation, so I assume you do most of the heavy lifting, but should users of this framework expect some cool hidden eggs when they look at their telemetry?