amangsingh

6 ms·

1.
by amangsingh 6mo ago
FaultWall sounds like we're absolutely on the same paranoid wavelength. Treating the LLM as a hostile/untrusted actor at the data plane is the only way this scales to enterprise. To answer your question: No, we actually completely
2.
by amangsingh 6mo ago
Spot on regarding the automod. Unfortunately, the way I naturally structure my writing almost always triggers a 50/50 flag on AI content detectors. It is the absolute bane of my existence. The filter instantly shadowbanned the Show HN
3.
by amangsingh 6mo ago
I looked into lat.md. They are definitely thinking in the same direction by using a CLI layer to govern the agent. The key difference is the state mechanism. They use markdown; I use an AES-encrypted SQLite database. Markdown is still just
4.
by amangsingh 6mo ago
To clarify the architecture: The LLM doesn't have access to the ledger. That’s the entire point of Castra. The LLM only has access to the CLI binary. The SQLite database is AES-256-CTR encrypted at rest. If an LLM (or a human) tries to
5.
by amangsingh 6mo ago
Thanks! This took a while (approximately 30 days) to get to this point. The market basically relies on two main alternative approaches right now, both of which have their merits: 1. File-based Memory (Markdown/Artifacts): Instead of j
6.
by amangsingh 6mo ago
Good question. I did think about behaviour trees early on, but I realized they optimize for the wrong thing in this specific domain. Behaviour trees are fantastic for agent autonomy; letting the agent dynamically construct its own path to a
7.
by amangsingh 6mo ago
Just dropped the Show HN here: https://news.ycombinator.com/item?id=47601608 . Would love to hear your thoughts on the architecture!
8.
by amangsingh 6mo ago
Just posted it here: https://news.ycombinator.com/item?id=47601608 Thank you so much for the coffee offer, that genuinely made my day! I don't have a sponsor link set up. Honestly, the best support is just hearing if t
9.
Show HN: Castra – Strip orchestration rights from your LLMs(github.com)
8 points by amangsingh 6mo ago|9 comments
10.
by amangsingh 6mo ago
Bingo. And them 'being careful' is exactly what bloats it to 500k lines. It's a ton of on-the-fly prompt engineering, context sanitizers, and probabilistic guardrails just to keep the vibes in check.
11.
by amangsingh 6mo ago
Claude code is a massively successful generator, I use it all the time, but it's not a governance layer. The fact that the industry is copying a 500k-line harness is the problem. We're automating security vulnerabilities at scale
12.
by amangsingh 6mo ago
Why not both? AI writes bloated spaghetti by default. The control plane needs to be human-written and rigid -> at least until the state machine is solid enough to dogfood itself. Then you can safely let the AI enhance the harness from wi
13.
by amangsingh 6mo ago
Herding cats is treating the LLM's context window as your state machine. You're constantly prompt-engineering it to remember the rules, hoping it doesn't hallucinate or silently drop constraints over a long session. System-le
14.
by amangsingh 6mo ago
If writing concise architectural analysis without the fluff makes me an AI, I'll take the complement. But no - just a tired Architect who has spent way too many hours staring at broken agent state loops haha.
15.
by amangsingh 6mo ago
A 500k line codebase for an agent CLI proves one thing: making a probabilistic LLM behave deterministically is a massive state-management nightmare. Right now, they're great for prompting simple sites/platforms but they break at l