hunterbrooks

5 ms·

1.
by hunterbrooks 5d ago
I think this is the key takeaway here, it's obviously a product for usage based customers
2.
by hunterbrooks 5d ago
You can do that on ellipsis.dev. Shoot me an email if you need help getting set up (email in bio)
3.
by hunterbrooks 5d ago
We do! Happy to get you set up, "schedule a demo" button goes to me: www.ellipsis.dev
4.
by hunterbrooks 5d ago
Awesome! I'm also working this, and looking for a cofounder. Feel free to shoot me an email
5.
by hunterbrooks 5d ago
The self hosted workers solve this use case. The control plane sits in OAI's cloud, but the actual tool calls are executed in your worker fleet. The main problem with this approach is that tool arg's get sent over the wire, and th
6.
by hunterbrooks 5d ago
Hi, can we schedule a call? Our product supports exactly this use case: www.ellipsis.dev
7.
by hunterbrooks 5d ago
How did you set all that up? We built a flow where you push a slack.yaml to github to arrange this
8.
by hunterbrooks 5d ago
> A competitor who is not an LLM lab gets their pick of the market at any given moment Exactly this. We're building an Agents-as-a-service API, would love to get your feedback: https://www.ellipsis.dev/docs/sess
9.
by hunterbrooks 25d ago
Doesn’t bother me at all. What’s the actual harm here? It’s arguably a better experience.
10.
by hunterbrooks 1mo ago
Lots of extra words in this blog post but the direction is right: companies prefer to build their own AI code review solutions than use off the shelf products. I’m the founder of a W24 company and we pivoted away from AI Code Review for thi
11.
by hunterbrooks 9mo ago
congrats Greg, Merrill, and the rest of the folks at Graphite! - Hunter @ Ellipsis
12.
by hunterbrooks 1y ago
Bingo - it’s colloquially
13.
by hunterbrooks 2y ago
Yes. My business is in the code review space, synthetic data is very helpful for evals
14.
by hunterbrooks 2y ago
LLM's get relatively better at read-heavy operations (ex: code review) than write-heavy operations (ex: code generation) as codebases become less idiomatic. I'm a cofounder at www.ellipsis.dev - we tried to build code generation f
15.
by hunterbrooks 2y ago
100ms is crazy fast. Approx what's the split b/t the embeddings calls and the graph traversal work (the library stuff)?
16.
by hunterbrooks 2y ago
Yes. Learning to "program" has never been easier, but nitty gritty stuff like networks and routing is still much more technical than the avg SWE
17.
How to move your GitHub App to Gitlab(ellipsis.dev)
6 points by hunterbrooks 2y ago|0 comments
18.
by hunterbrooks 2y ago
HA! Your LLM agents/prompts nailed a hilarious tone. For https://ellipsis.dev > This website looks like it was designed by someone who just discovered the concept of "AI" and decided to slap it all over a half-
19.
by hunterbrooks 2y ago
Nice, I imagine you're using your own product? I got "LAUNCH SALE $100 OFF" 1. How do you provide consistent banners to an individual? or a device? 2. Are you aware of anyone doing this for Saas products?
20.
by hunterbrooks 2y ago
Cool, probably not relevant info for modern devs because of all the hosted git providers, but interesting nonetheless. Got any thoughts on how LLM powered code generation products can use format-patch? Also, 2009 was pre git?!
21.
by hunterbrooks 2y ago
Pricing is on the landing page. $20/seat per month
22.
by hunterbrooks 2y ago
Good catch. The team could add this rule to their Ellipsis config file to make sure that it's always flagged: "Never use magic numbers. Always store the number in a variable and use the variable instead." Docs: https:/&
23.
by hunterbrooks 2y ago
Unfortunately not. We’re early - all our data is qualitative. But we know we need hard numbers, so we’re working on it. We don’t want to sell a novelty, the product needs to actually save time
24.
by hunterbrooks 2y ago
No way, anyone can make silly mistakes. Ellipsis is a 2nd pair of eyes to catch stuff like that
25.
by hunterbrooks 2y ago
The product works on draft PR's too, but not on local. Sometimes reviewers rope in Ellipsis by asking questions (we also support natural language Q&A about a PR) or by having a design discussion via GH comments and then assigning t
26.
by hunterbrooks 2y ago
As it stands today, Ellipsis isn't sold as an AI software engineer. One of our largest learnings is that state of the art LLM's aren't good enough to write code autonomously, but they are good enough to be helpful during code
27.
by hunterbrooks 2y ago
Automated code reviews are a step in the software development lifecycle, not a replacement for human reviewers. Ellipsis can't commit code without your permission and approval, so this particular parrot can't feather your filesyst
28.
by hunterbrooks 2y ago
All those tools do code review, so you'll have to try them out for yourself to see which is the most helpful. But when it comes to writing code for you, not all those tools actually run your unit tests/linter/compile/etc
29.
by hunterbrooks 2y ago
Hmm, that searches issues, which isn't the best way to see Ellipsis' work. Example of PR review: https://github.com/getzep/zep-js/pull/67#discussion_r1594781... Example of issue-to-PR: https:/
30.
by hunterbrooks 2y ago
Ellipsis increases the quality of code a developer brings to their teammates, meaning fewer PR comments, meaning the team ships faster. It's not a replacement for human review. It's most often used by the PR author after a self re

More ›