theo31

7 ms·

1.
Show HN: AgentCloud – MCP-first iOS simulator cloud
1 points by theo31 21d ago|0 comments
2.
We built secure automated learning loops in Modal and Claude Code(twitter.com)
2 points by theo31 3mo ago|0 comments
3.
by theo31 1y ago
I use it and I love checking it in the morning to see what my team is up to, I don’t have to ping people and break their flow. It helps us figure out what got done and where we are in our roadmap
4.
Show HN: Inferrd – open-source ML Deployment(github.com)
4 points by theo31 5y ago|0 comments
5.
Show HN: Deploy any ML models with 1 line(inferrd.com)
1 points by theo31 5y ago|0 comments
6.
Show HN: GPU-Accelerated Inference Hosting(inferrd.com)
2 points by theo31 5y ago|0 comments
7.
Show HN: Deploy Scikit, TF and SpaCy on GPUs in 2 Minutes(inferrd.com)
2 points by theo31 5y ago|0 comments
8.
by theo31 5y ago
Those frameworks are installed by default in our custom environment. There is no additional setup/configuration required from you.
9.
by theo31 5y ago
At the moment, no, only the random hash gives some kind of security by obfuscation. More advanced security controls are coming soon.
10.
by theo31 5y ago
We don't use TensorRT at the moment, but it is something that we are exploring.
11.
by theo31 5y ago
Oh that's very interesting, how ready for production is it? It only works for TF right? > If you need a few dozen inferences per second per server, this is the cheapest way. And you're not depending on a proprietary solution wh
12.
by theo31 5y ago
Sorry that's a typo, they are K80s: https://www.nvidia.com/en-gb/data-center/tesla-k80/
13.
by theo31 5y ago
Yes! A more in-depth blog post is coming soon. We do host the hardware ourselves, for complete control over the GPUs. We found a great infrastructure provider that is also experiencing shortages.
14.
by theo31 5y ago
The response time guaranteed is for a reasonably sized model. Bigger models (> 700 MB) will take a bit longer. The model size is the zipped size of your model that is uploaded to Inferrd (either through the SDK or the website). I'll
15.
by theo31 5y ago
Yes you absolutely can!
16.
by theo31 5y ago
Thank you! We don't have any cold start delay! In our custom environment, you can do exactly what you are describing (running both CPU and GPU code). We provide you with access to the GPU and the CUDA libraries installed. It's bas
17.
by theo31 5y ago
Yes absolutely, you can run almost anything in our custom environment.
18.
by theo31 5y ago
There is no cold start! We keep your service hot all the time.
19.
Show HN: GPU-Accelerated Inference Hosting(inferrd.com)
56 points by theo31 5y ago|34 comments
20.
How to Save and Deploy TensorFlow(inferrd.com)
1 points by theo31 5y ago|0 comments
21.
Show HN: Deploy Keras without any configuration(inferrd.com)
2 points by theo31 5y ago|0 comments
22.
Show HN: The easiest way to deploy XGBoost(inferrd.com)
1 points by theo31 5y ago|0 comments
23.
Deploy Scikit with Flask and Docker (Tutorial)(inferrd.com)
2 points by theo31 5y ago|0 comments
24.
Show HN: Deploy Scikit to the web in 2 lines(inferrd.com)
1 points by theo31 5y ago|0 comments
25.
Show HN: Deploy TensorFlow with just 2 lines(inferrd.com)
2 points by theo31 5y ago|0 comments
26.
Deploy Scikit Models from DeepNote on Inferrd.com(inferrd.com)
1 points by theo31 5y ago|0 comments
27.
Show HN: Deploy ML with Ease(inferrd.com)
1 points by theo31 5y ago|0 comments
28.
by theo31 5y ago
Definitely not a CTF challenge. The product is still very early on and a lot of features are hidden/hard to find (that's one of the reasons the signup isn't open). I'll send you some documentation soon.
29.
by theo31 5y ago
That's fair, I'll come back with something to show :)
30.
by theo31 5y ago
It's because I like working closely with early users to nail the features and experience instead of opening the floodgates right away.

More ›