davidvgilmore

7 ms·

1.
by davidvgilmore 3mo ago
For those that have capable enough hardware, it's effectively free to run subtasks on-device. (just the marginal cost of additional electricity). With Google's most recent 12b param Gemma model, even Mac users with just 16gb of un
2.
by davidvgilmore 3mo ago
Many of them are produced by Chinese labs. Some, like Neomotron, are U.S. made. And we support inference providers in both the U.S. and overseas. If geography is important, we can restrict which geos inference takes place in. And if you don
3.
by davidvgilmore 3mo ago
Four ways: (1) We are built specifically for Claude Code model routing. (2) We route at a subagent/subtask level. (3) We support on-device routing. (4) We have a built-in ML router trained specifically to route Claude Code subagent tas
4.
by davidvgilmore 3mo ago
Yes - in short, open models like Deepseek, Mimo, Kimi, and GLM tend to complete tasks with less tokens and cost less per token than both Sonnet and Haiku. So those models are more cost efficient, and we often think of that as them having hi
5.
Show HN: Rayline routes Claude Code subagents to on-device and cheaper models(rayline.ai)
11 points by davidvgilmore 3mo ago|9 comments
6.
Show HN: Dbt-skillz compiles your dbt project into a Claude Code skill(github.com)
4 points by davidvgilmore 6mo ago|0 comments
7.
Takeaways from 10k vibe-coded projects(dvg.blog)
4 points by davidvgilmore 1y ago|0 comments
8.
Sales-led entrepreneurship (w/ $100m exit) to a product-led startup is humbling(dvg.blog)
1 points by davidvgilmore 1y ago|0 comments
9.
by davidvgilmore 1y ago
Your comment is fair. I don't love vibe coding as a category name. And somehow using vibe coding to describe something whose namesake, the Memex, had such virtuous ambitions feels cheap. But I ALSO want people to understand what our pr
10.
by davidvgilmore 1y ago
Memex makes it easy to ideate, research, and build projects without writing a line of code. It's a fully chat based interface (not an IDE). Its GUI allows you to visualize data inline, perform deep research-esque queries, and create +
11.
by davidvgilmore 1y ago
HN Community: We had this clause in our Terms and Conditions: "you will not (and will not enable others to) use the AI Features: (e) for the development of any service or other offering that competes with or replicates the Services.” T
12.
by davidvgilmore 1y ago
HN Community: We had this clause in our Terms and Conditions: "you will not (and will not enable others to) use the AI Features: (e) for the development of any service or other offering that competes with or replicates the Services.” T
13.
by davidvgilmore 1y ago
thanks!
14.
by davidvgilmore 1y ago
Fair
15.
by davidvgilmore 1y ago
That's fair. We do plan to open source components of Memex to allow for extensibility while also having a strong enough revenue model to build without being fully dependent on VCs
16.
by davidvgilmore 1y ago
That's a good feature idea and something we've considered but not gotten around to yet. We do have a secrets mgmt feature that uses keyring, so you can store secrets through the app in your system keychain, which then requires you
17.
by davidvgilmore 1y ago
I suppose you copy and paste this for each startup launch you see?
18.
by davidvgilmore 1y ago
Hi @thro1 - this is my first post to the HN community. Memex is something I and my cofounders have worked hard on. While I expected some trolling, your comment is just baffling. To anyone reading this exchange: I wrote the Show HN attemptin
19.
by davidvgilmore 1y ago
in this context, not opionionated == you can use it to build with Rails, FastAPI, Firebase, python + modal ... whatever tech stack you want to use. Some will work better than others just due to the nature of the LLMs. We've heard loud
20.
by davidvgilmore 1y ago
Honestly I hear you on this one. But the market has really taken up "vibe coding", so it's been the easiest way to clearly communicate to people what Memex does.
21.
by davidvgilmore 1y ago
thanks - yeah, we've heard this feedback loud and clear from the HN community. We're cooking
22.
by davidvgilmore 1y ago
thanks - yeah, we've heard this feedback loud and clear from the HN community. We're cooking
23.
by davidvgilmore 1y ago
In general, we can't see what users are doing. But we can see some things like that they upgrade to new releases. We only site users @ logos that are using it on a consistent basis
24.
by davidvgilmore 1y ago
we can't see their activity other than that they have accounts and have used it. In the case of Google, we've had folks from both PM & engineering. We've talked with our PM user who has been using for prototyping.
25.
by davidvgilmore 1y ago
thanks for calling this out To be 100% clear: we do not prevent anyone from using Memex to build a Memex competitor. That clause was boilerplate our lawyers included, and it doesn't reflect how we operate or what we intend to enforce.
26.
by davidvgilmore 1y ago
thanks - great feedback on (1) re: multi tasking ... that's a known UX limitation. You can "command+click" the logo in the top left to open a new window, then just keep the convos open. That's the way to multi-task with
27.
by davidvgilmore 1y ago
Our namesake is Vannevar Bush's "Memex": https://www.theatlantic.com/magazine/archive/1945/07/as-we-m... We like the ethos of peaceful scientific progress. And we're most passionate a
28.
by davidvgilmore 1y ago
We have users at Google that use Memex
29.
by davidvgilmore 1y ago
You can use it for paid apps, open source apps. Whatever you want. Everything you create with Memex is 100% yours. That clause is only meant to keep people from reselling our app, Memex, itself. But I agree it's super confusingly writt
30.
by davidvgilmore 1y ago
The main difference is that Memex isn't an IDE. It's a chat-based, full agentic coding experience. You can see the code it creates, but you can't edit it (yet). It's not for everyone, but as agentic coding gets better we

More ›