myprotegeai

10 ms·

31.
by myprotegeai 2y ago
I wonder if it has anything to do with what is going on in my /var/log/syslog. I have hundreds of entries of the following: 2024-09-20T13:28:42.946055-07:00 hostname kernel: audit: type=1400 audit(1726864122.944:11828880): ap
32.
by myprotegeai 2y ago
There is no "standard way", only rough principles, and it all depends on the unique DNA of the company (cloud, stack, app deployment targets, etc). Yes there are a lot of tools, and experienced infrastructure engineers spend a lot
33.
by myprotegeai 2y ago
This topic lacks nuance. I agree in focusing on building things that people want, as well as iterating and shipping fast. But guess what? Shipping fast without breaking things requires a lot of infrastructure. Tests are infrastructure. CI
34.
by myprotegeai 2y ago
A single VM is all fine and well until your hacky go-fast code allows an issue with a single request to take down your service. Serverless requests are isolated and will limit the blast radius of your hacky code as you iterate quickly.
35.
Function-Level LLM Integration(github.com)
2 points by myprotegeai 2y ago|0 comments
36.
by myprotegeai 2y ago
A python library for calling an LLM like calling a function https://github.com/amoffat/manifest
37.
by myprotegeai 2y ago
CaaS - Collusion as a Service
38.
by myprotegeai 2y ago
And OpenAI also has Structured Outputs[1] that has the same effect as Guidance. I use it to safely deserialize remote function calls based on a jsonschema[2]. It works very well. 1. https://openai.com/index/introducing-
39.
Are we in an AI hype cycle? [video](youtube.com)
1 points by myprotegeai 2y ago|0 comments
40.
by myprotegeai 2y ago
I send dynamic jsonschemas to LLMs in manifest[1] to simulate function calls based on function signatures. These structured outputs will be enormously useful. Other posts claim that you can generate jsonschema-conformant output reliably wit
41.
Choose your own adventure in 100 lines with micro agents(github.com)
3 points by myprotegeai 2y ago|0 comments
42.
VLLM hosts local LLMs easily(docs.vllm.ai)
2 points by myprotegeai 2y ago|0 comments
43.
by myprotegeai 2y ago
How many Farmbots would I need to feed 2 adults and 2 children year round? What challenges might I run into? Edit: Found a link to yield analysis https://farm.bot/pages/yield It appears the answer is however many Farmb
44.
Show HN: AI-Manifested Code Execution(github.com)
1 points by myprotegeai 2y ago|0 comments