phil

ref: language design 
(In the era of →  LLMs)

ref: language design 
(In the era of →  LLMs)

https://youtu.be/MUol_5xvFTI?si=cNxTetXe1lBgUT0y

Anders: "Why not AI" 
Deterministic requirements of the TS → Go port project.
timestamp

🐵 ← 🙈🙉🙊 → /phil →  phil.cockfield.net, db.team →  
ƒ(🌳) →  🧫 /sys  ←  @cell <T> 🧠🐦‍⬛, 100% human cartoon

Anders: "Why not AI" 
Deterministic requirements of the TS → Go port project.
timestamp
https://youtu.be/UJfF3-13aFo?si=6XbfPu3Efp1-FOdb&t=1071

Check out the excellent YouTube video shared by @pjc where two engineers discuss about the future of programming languages, AI, and developers.

What has been more interesting to me is that people with really good skills in writing and reading languages (not only programming languages) are going to excel in the coming years. Why? Ambiguity is the worst part of LLMs. Being concise and precise in requesting information and giving instructions is a skill that not many people have. 

I think engineers, due to our training, have a more structured head regarding ambiguities, so we avoid ambiguous words to transmit the message. However, reading and understanding books is rapidly disappearing due to LLMs making nice voice summaries for them.

I also loved that we might end up generating specific DSLs for coding with AI so we can get specs properly passed to machines (hello, MCP, even though you are not a DSL per se) and get reproducibility and stability back, something that we all seek in any digital product.

cross-ref: https://farcaster.xyz/teleyinex.eth/0x05ca4228
via Daniel ← @teleyinex.eth

🌷💀⏳

teleyinex.eth ▫️

teleyinex.lens ▫️

teleyine.x ▫️

daniellombrana.es  ▫️

@astaralabs ▫️

@obeygiant ▫️

@scifabric ▫️

dev

Dad, geek, conversation liquidity provider. vrypan.net (home), @l--o--l (fun). devcoin:$lemon3