3,340 karma · joined October 3, 2010
- Helped start + Board @ CharmIndustrial.com
- Co-Founder @ Kradle.ai
- Co-Founder/CEO at Firebase (YC S11)
- Co-Founder @ CovidActNow.org
- Partner @ HF0
- Built the product team @ ChainlinkOn one walk I asked 'why are you doing another startup?'
For context, his last one, OpenSea, was valued well into the billions, so it wasn't for money.
His reply: "I just love solving all the puzzles."
It's incredibly hard to compete with someone who is playing the game for the love of the game.
Kudos on playing well, Alex.
Seeing traces in real-time (and Claude seeing them too) is so so good. The velocity speed up is hard to understate.
Request for the Raindrop team: I wish your eval support was better. It’s not great having it completely disconnected from my CI system.
Though, their console feels like it didn't get the love that the rest of the infra / website did.
Congrats on the 1.0 launch! I'm excited to keep building with Instant.
I loved this place as a kid and would marvel at the patterns.
The air of mystery around who made it and why invites some fun speculation.
Kudos Tasklet.
We've built our infra so that we can plug in any simulation environment. If an AI-generated world starts being programmatically modifiable (and has really solid object permanance :) then we'd happily use it!
Minecraft has a DSL that lets you manipulate the world. We've piggy-backed on that, along with a K8s infra to run N worlds in parallel, to let you create simulations of arbitrary complexity.
We think simulations are the best way to test frontier AIs due to their degrees of freedom and expressivity.
AMA!
- https://the-inference.vercel.app
- https://github.com/jamestamplin/instant-testThe experience was brilliant.
Pros:
+ Fast
+ Easy
+ "Vibe coding on steroids" basically
+ The sense of 'wow' that comes very rarely with new tech
Cons:
- It used Instant as the database/backend, but I wasn't sure what it had done / how exactly it worked and had to spend a bunch of time asking Claude + reading the code to get it. It seemed reasonable, but if I were doing a prod system vs a PoC, this is where the time would be spent. ("Vibe coding lets you create tech debt 10x faster")
Net-net: This is the way for prototyping / validating. This is probably the way for production systems in N months too once the toolchain + agents get better.
Good luck, Joe, Stopa and team!
We were previously working on a chat system called Envolve (https://www.envolve.com), that was 'Facebook Chat for any website'. A game that was using us for in-game chat created channels, used display: none on them, and passed game state through the chat.
We scratched our head, asked them why, and learned they wanted to focus on the frontend, not to deal with realtime message passing.
This led us to create a 'headless version' of our chat infra (re-written in Scala) that became the Firebase Realtime Database.
There was a Titan Bluetooth Key (for 2FA) Vulnerability, you've said you'll replace the affected keys[1], but you're no longer doing so. Which is frustrating.
[1] https://security.googleblog.com/2019/05/titan-keys-update.ht...
I no longer work at Firebase / Google, but two points:
1. There may be issues with the GCP integrations & UX/DX, but GCP integration is good for many customers and necessary for the future of the business.
One of the common failure modes for the 2011-2014 crop of Backend-as-a-Service offerings was their inability to technically support large customers. The economics of developer tooling are a super-power-law. So, if you hope to pay your employees you'll need to grow with your biggest customers.
Eventually, as they become TheNextBigThing, your biggest customers end up wanting the bells and whistles that only a Big Cloud Platform provide.
This was a part of the reason we chose to join Google, and why the Firebase team really really really pushed hard to integrate with GCP at a project, billing, and product level (philosophy: Firebase exposed the product from the client, GCP from the server) despite all the organizational/political overhead.
2. I'm excited to see the current crop of app platforms emerge. It has been 10 years since we launched (https://news.ycombinator.com/item?id=3832877) and there are now some great innovations in the space. I like the way Supabase (https://supabase.com/) has exposed Postgres and InstantDB (https://www.instantdb.com) graphdb+realtime is really promising.
Most of the credit goes to my co-founder, Andrew, and the whole team who gave nothing short of everything to build it.
Eg. Fabric’s products were rolled into Firebase with continuity for developers. AngularJS was replaced by Angular.
Yes, there's a trend of investors seeking to deploy capital towards ESG (thankfully!)
No, this isn't going to gussy up pure-tech with the 'climate' label. I say this as former-USV founder who is now a USV LP. I can also attest to how solid the humans at USV are.
[1] https://www.investopedia.com/terms/e/environmental-social-an...
I visited two of Ackerman's caves (Spring Valley and Temple of Doom) in May of 2002. It was a few days shy of high school graduation and summer was starting to arrive in Minnesota.
My future Firebase co-founder, Andrew, and our high school classmate, Bob, made the drive down there one Saturday. Bob had been a sort-of pupil of Ackerman's, helping him explore and map the various caves he had found.
The day started innocuously enough with a visit to the main Spring Valley Cave that the article references as being previously owned and equipped to give tours. Bob took us off the steel gantry-ways and into some side passageways and chambers that were rarely trafficked.
It was the second cave, Temple of Doom, that was the more, ah, interesting experience. It has only recently been discovered and wasn't fully mapped yet. We popped open a man-hole cover in the middle of a corn field and dropped down a ~50 foot steel ladder. From there the cave descended in a spiral, with the passageway barely able to fit a human. Bob, typically the most composed human I knew, occasionally burst into fits of expletives, followed by silence. "Uhhh... Bob, everything OK?". Descending a 20-30 degree downward slope with inches to spare on every side, occasionally getting stuck, and with no way to turn-around was hair-raising.
After some undetermined period of time we finally got to the bottom of the spiral. At this point, Bob announced we could either brace ourselves between two sheer, very wet, rock faces and climb up ~100 feet without ropes, or go back the way we came. So... up the rock faces we went. Fortunately nobody slipped.
I can still viscerally feel the elation of emerging into bright afternoon sun of the corn field, grateful and relieved to be uninjured.
This was one of the several intense and bonding experiences Andrew and I had together, and perhaps contributed to why we kept working together despite some early startup failures.
As for Bob, he kept on spending time with Ackerman, until one day he was buried alive by a cave entrance collapse. Ackerman dug him out with a backhoe, but he was a little more wary of caving after.
The model is run (and so forward projections are updated) once every 3 days at the moment.
Our model is only scoped to the USA (I wish we could include more countries, as this is a global pandemic. We haven’t had the time to expand yet).
I’m sure the API could be improved in the next version. For example, we had some lively discussion around whether OBSERVED_INTERVENTION was intuitive enough variable name. (It uses an inferred R0 based on actual observed deaths, cases, and hospitalizations to generate forward projections).
If you have any feedback, let us know and we’ll try and make it better in V2.
They're based in LA & have called their approach CPS - Camera Positioning System.
https://www.stuff.co.nz/travel/themes/adventure/103873808/sv...
Crunchbase: https://www.crunchbase.com/organization/bebop#section-overvi...
Previous HN discussion: https://news.ycombinator.com/item?id=10839646