I find the #claudecode source code interesting.
-
I find the #claudecode source code interesting.
It's not that different from codex or gemini. Shares many similarities.
Good job.
-
I find the #claudecode source code interesting.
It's not that different from codex or gemini. Shares many similarities.
Good job.
The "adversarial verifier" is key, imho.
I see it quite often, and it kicks in for complex challenges. I didn't know you could do that.
-
The "adversarial verifier" is key, imho.
I see it quite often, and it kicks in for complex challenges. I didn't know you could do that.
Context pruning, minimum complexity, and faithful reporting are key things to consider.
Alright. That makes total sense.
-
Context pruning, minimum complexity, and faithful reporting are key things to consider.
Alright. That makes total sense.
Ahh you don't stdin the code base. You segment it into pieces. It reads code, rather than copying it over. The client assumes that it cannot just send over x thousand lines and ask for inference.
Of course...
-
Ahh you don't stdin the code base. You segment it into pieces. It reads code, rather than copying it over. The client assumes that it cannot just send over x thousand lines and ask for inference.
Of course...
They use Zod schemas against hallucinations.
They should write books. This is cool.
-
They use Zod schemas against hallucinations.
They should write books. This is cool.
There is a semantic firewall against hallucinations.
Of course... they use an agentic pattern for that.
-
There is a semantic firewall against hallucinations.
Of course... they use an agentic pattern for that.
I don't think this was a marketing gig leak. There is too much IP in there.
People have no idea how difficult this stuff can be.
-
R relay@relay.infosec.exchange shared this topic
