TheFirstNubian

Saw this earlier on LinkedIn and checked it out. Awesome initiative!

Location: Accra, Ghana

Remote: Yes (or hybrid)

Willing to relocate: Yes

Technologies:

    - Core: Rust, C/C++, Solidity, PostgreSQL, WebAssembly

    - Domain: Zero Knowledge Proofs (ZKPs), Solana, compilers

    - Tools/Libraries: Winterfell, Arkworks, RISC0 zkVM, Solana Anchor

Email: kobby5angels@gmail.com

LinkedIn: https://www.linkedin.com/in/kobbypentangeli/

GitHub: https://github.com/kobby-pentangeli

Rust/C++ cryptography engineer with strong bias towards zero-knowledge proofs (ZKPs) and compilers. Currently leading the open-source implementation of a Turing-complete programming language for writing ZKPs; Rust-native syntax, Winterfell STARK backend: https://github.com/maatlabs/maat

The bit about treating one’s prompt history as a legal document has really struck a nerve with me. I’ve been keeping a separate git history solely for my prompts. Initially, the goals were simple: reuse prompts, turn some into skills, etc. But in light of the insights from the article and the discussions here, I need to treat this practice as serious business.

The elephant in the room, of course, is what constitutes “meaningful human authorship.” However, I cannot shake off the feeling that all user interactions with these AI models are being logged. Perhaps this may turn out to be the bigger concern in a potential legal battle than code authorship.

That seems to be the general direction, at least from my daily dose of cope on X (Twitter). Regardless, conscious design will never go out of style.

Lol! Wrong choice of word, maybe. I meant to say that we don’t seem to be putting much thought into how we’re outsourcing thinking to the LLMs.

I’ve used Rust daily for the last 6 years; from writing compilers and WASM runtimes to writing smart contracts and SDKs… and the experience has been nothing short of amazing.

I’m not sure what you mean by a “casual setup,” but I find Cargo to be very easy to use. Granted, compilation can be pretty slow, but it’s a small price to pay for the runtime performance.

If by “faster feedback” you mean something that looks and feels like Python’s REPL, then I think you’re missing the point here. Rust is a compiled language.

I’m a little conflicted on this, as I see a slippery slope here. LLMs in their current state (e.g., Opus-4.7) are really good in planning and one-shot codegen, which I believe is their primary use case. So they do provide enough leverage in that regard.

With this new workflow, however, we should, uncompromisingly, steer the entire code review process. The danger here, the “slippery slope,” is that we’re constantly craving for more intelligent models so we can somehow outsource the review to them as well. We may be subconsciously engineering ourselves into obsolescence.