Learn about AI, Agents, or AI Benchmarking! I've been an engineer for more than 20 years and absolutely love technology, and have become absolutely consumed with learning everything I can about AI.


GosuCoder

Augment Code recently released their remote agents, they kindly offered the community 4 free subscriptions of 1 month. I setup a google form to capture entry, only the emails of the winners will be passed on to Augment code so they can activate the trial on your account.

Its definitely worth entering if you haven't tried out remote coding agents yet!

Drawing on June 8th!

docs.google.com/forms/d/e/1FAIpQLSfPXVtoUQLrGWsyl1…

6 months ago | [YT] | 26

GosuCoder

Google just announced Gemini Code Assist! blog.google/technology/developers/gemini-code-assi…

Very curious to see how this compares!

9 months ago | [YT] | 10

GosuCoder

Claude Code eats tokens like you wouldn't believe! 2 hours of using it, already at 7.6 million tokens in, and 330k tokens output. Its really awesome though.

9 months ago | [YT] | 20

GosuCoder

Oh yes!!! Claude 3.7, should I run some tests on this against Grok 3 Think tonight? I could Live stream or put together a video, what do you think?

9 months ago | [YT] | 23

GosuCoder

First Tests with Grok 3 have not been super great at coding for me, BUT I don't have the Big Brain/Thinking version yet, so I'm hoping that will make a big difference. I wish I could attach short video's to posts.

10 months ago | [YT] | 8

GosuCoder

The worst feeling in the world is when something is broken in code, and you debug and debug, and then all of the sudden it starts working again. Literally changed nothing and just started working. Most likely some sort of cache issue right? right?

10 months ago | [YT] | 10

GosuCoder

RTX 5090 is definitely a paper launch, in line exactly as it opened up, and instantly errored. Eventually ended up in line to be told its out of stock. Same story on NewEgg. Were any of you able to snag one?

10 months ago | [YT] | 2

GosuCoder

Qwen 2.5 VL - 7B is testing really well. Here is an example of looking at the NVidia stock chart. Absolutely insane to see this working so well in a small model. Prompt: "Describe the details of this chart"

I've also included a subset of the response Qwen2.5 VL - 7B has on the DeepSeek R1 Reasoning benchmarks. I can't easily fit the entire results into an image, but for a 7B model this is INSANE!

10 months ago | [YT] | 13