i recently used Magic Patterns for a very niche use case and had a great experience:
i wanted to do show new customers examples of how they can use my product, which lives primarily in email.
to do it via Loom I would need to create tons of fake email addresses and juggle a whole complicated set of scenerios. and to do it in after effects would take forever.
Yes, it's incredible seeing what you can do with pure code. I think that's a lot of the magic here: with code, you can do anything. This part is super gratifying to work on.
From day 1, we have thought of code as a first-class citizen in Magic Patterns because that's all the LLM sees. So then at the application layer, it becomes our job to best help the user interact the LLM a.k.a feed it the most relevant code. This part is suuuuuper challenging.
The spike in interest from HN has been totally overwhelming, so while we are thrilled with the validation of this concept, we are running out of stock and falling behind on delivery schedule.
I'll be sure to reach out individually to everyone who submitted an order.
We can't wait to launch this service in a way that can handle demand spikes like this -- thank you all for yor support.
Edit: for reference, we've had orders for 150 loaves of bread in 2 hours.
So you achieved your goal right? Historically bakeries are very 'spikey' in their demand, a friend of mine runs one and they spend 3AM to 6AM making donuts and bagels which they sell out of between 6AM and 7:15AM, and then sell maybe a handful or two for the rest of the day.
Given the complexities of bread, it would interesting to hear a post mortem about ways you might both account for spikes without compromising delivery times, delivery scheduling to make maximum use of delivery resources, and even recipe variations to support those goals.
That's an interesting logistic problem I hadn't even thought of. I bake a lot of bread (today's was buttermilk chile cheddar) and it can be very uncooperative and unpredictable, unlike cakes and other pastry.
I understand that in a production environment they will be more process driven than my own kitchen, but I wonder what the timing variation is in the typical small commercial bakery. i.e., from start of mixing to fresh bread out the oven, how repeatable is the time to completion?
There is a donut place in Berkeley that works very similar to that. The reason for them is that there are a lot of large-scale orders that want the donuts for breakfast (e.g. the school dining facilities).
150 tiny orders is like cutting down a forest with a herring. Catering events is where the real cash is at. Large universities like Stanford operate like cities with 100s of autonomous departments free to choose whichever vendor/s they like.
> Beyond bread, we offer a rotating selection of coffees, cheeses, cured meats, oils and vinegars and more. Each is locally produced and carefully selected.
Actually in the modern business world, as long as you're growing, you're survivable. Cash influxes (loans, investment) can support a business for years - Amazon did it. You just need to convince investors that you have a story.
That's an anecdotal example. Amazon is different in that Bezos has enough room to dial back reinvestment and generate more cash or turn up reinvestment to reduce apparent profit.
Profit doesn't just magically appear, costs and pricing have to eventually face reality. Sooner, the better.
Spending more money does not necessarily make a business any better. That's like these noob business people that buy stock before they sell any of it... An example of a horrible waste of cash.
Without profit, it's not a business, it's an idea-guy dream.
No that's just an 'example'. Many startups fit this model. I'm in one that's in the 5th year, still growing, still with financial backing. Many are in this mode until bought out (or of course until they fail).
They didn't just get a lot of people to their page. They also got a large chunk of their visitors to pay them. The first part might not be sustainable - and it seems they weren't planning for it to be as this was a one time offer - but it definitely validated something, and also put them in touch with a lot of customers.
"so while we are thrilled with the validation of this concept"
When you say "validating the concept" what do you mean? That if you get a high amount of orders you can deliver?
Because obviously you can't depend on getting placement on HN or equivalent to drive business. And the cost of customer acquisition could kill your model assuming the price point stays the same.
For me, ready-to-eat health foods. However, healthy is subjective in a world where low-fat, high-carbohydrate diets are promoted by the American Heart Association. I'm not paleo, but it's the best litmus test I use for if food is healthy.
I don't want to discount what you guys/gals are doing. I think this is great and don't let my one data point detract from your mission. There are tons of bread lover's out there, unfortunately I'm just not one of them.
from time to time doesn't work for people who have full-time gluten intolerance. you need to either support it fully or don't bother diluting your brand.
Well, TaskRabbit could do it at a much higher price.
Our thinking is that if you like high quality, fresh bread, then you can sign up to each week have options of different bread from different bakers – and when you see one you like, we'll deliver it to you.
By offering just one product each day, we're able to do deliveries highly efficiently (like newspaper delivery, not pizza delivery), so we're able to deliver a loaf of bread to your house for a price that compares to what you'd pay at the bakery or a market that sells high quality bread.
Newspaper delivery is cheaper than a newstand. I hope you can get to the scale where daily delivery with a six-months subscription, for example, is even cheaper than going to a bakery.
If you think your only competition is Wonderbread, then that makes sense. However, there are places like Trader Joe's and Whole Foods that offer bakery fresh high quality bread that isn't "tough to find" - they're everywhere.
We're still working hard on finishing up the tech behind this, as well as locking in partnerships with a few more bakeries before we launch with a small group of early customers, but I wanted to post here and gauge interest among the HN folks in SF.
We have about 300 people on our email list so far, and will try to keep our launch small so we can deliver top-notch service, but please sign up if you are interested in this.
The reason we're so excited about this project can be boiled down to logistics. To simplify and streamline logistics, our business is modeled more closely to the milkman than the pizza delivery man. On Sunday, we send out a picture message with one bread choice for each day of the coming week. If you want any of them, you reply via SMS and we bring it to your doorstep.
Since our drivers can make one pickup and then drop off identical products to every customer, delivery becomes far simpler and cheaper.
Lots of smart folks are vying for the one-stop grocery delivery market, and we think it will become a war of logistics for who wins that massive market (personally, I'm pulling for Instacart). We're approaching a different and smaller (but still quite large) market, from a different angle.
Fresh bread is our first product because it's almost universally loved, and it arguably decreases in quality every hour after being baked. Eventually we hope to expand to still-simple, curated line-up of high quality, short shelf-life specialty foods (coffee, cheese, etc.)
Sign up if you're interested, but more importantly, leave us some feedback here. :)
It was common to have your milk delivered every morning until a few decades ago(in the UK at last), with some milk men also doing eggs, butter, orange juice etc.
They have mostly died out. Have you explored why and what you will do to succeed when these people have closed their businesses?
You should list some of the bakers you get bread from on the site/signup and not just in the app. I'm hesitating signing up because I have no real interest in bread delivery unless it's from certain bakeries that I know are worth the extra effort over buying a loaf of Acme bread or whatever at the local Safeway or corner market. Like Josey Baker and Arizmendi, which I see in the screenshots. And even then it's like well... I don't just want bread, I also want like a bunch of their other products (scones, croissants, etc.). Maybe I'm better served by Postmates/Taskrabbit, but more bread products than just loaves of bread would be awesome albeit more complicated.
Also if you could deliver Tartine bread I would be so all over that in an instant! Especially right out of the oven :) Even though I like to bake and I have two loaves' worth of dough proofing in my kitchen right now.
I would be willing to risk $6 one time to test out the service. If the bread wows me, I would not care which bakery the bread came from. If it underwhelms, I have not lost that much.
It's not the $6+ or the service as much as it is my existing bread preferences. Like if all they did was offer Firebrand bread you couldn't pay me to buy that or spend my time even looking at the service, whereas I would pay $15 for a loaf of Tartine delivered within a couple hours in a heartbeat.
Maybe it's because I have such a high standard for bread at my table (my roommate is a foodie too and I can go to Thorough Bread & Pastry, Bi-Rite Market, or Tartine in 10 minutes... or I can bake my own sour sourdough in 30) but knowing where the bread comes from interests me more than what varieties are offered or how convenient the service is.
But I say this as someone that just baked a loaf of beer+cheddar+mustard bread and I smell like pure deliciousness that you can't buy from a delivery service, so maybe the service isn't for me.
i wanted to do show new customers examples of how they can use my product, which lives primarily in email.
to do it via Loom I would need to create tons of fake email addresses and juggle a whole complicated set of scenerios. and to do it in after effects would take forever.
so i used magic patterns to make an app that lets me upload JSON scripts of the email threads, and it animates them. if you skip to ~1 min mark on this video you can see the output https://drive.google.com/file/d/1iWC5U2Q3x30I5m1bTuN9c2OnfDo...