if you left software engineering in 2024 and came back in 2026, you would've missed:
MCP/A2A
RAG
Proompt Engineering
AI tab complete
So basically not much because that's all irrelevant now
Here's my AI coding workflow and all the skills I'm using:
Idea -> /write-a-prd -> PRD
PRD -> /prd-to-issues -> Kanban Board
Kanban -> ralph.sh -> Ralph Loop
Ralph Loop -> Manual QA
Links below to skills
@proxy_vector@zeeg disagree, libraries with >1k usually mean someone solved a problem for you , no need to reinvent the wheel. ive seen so many projects fail or delay because of stubborn devs proving they can diy react-query and then leave the projects.
@Jeanvaljean689 i feel this will change with AI disruptions. linear work and knowledge work most likely wont be very relevant in the coming years and adhd folks can benefit from that.
@sqs@pierrecomputer hello sir, i was told using ampcode is a negative signal :/ the headline was like “if you are still using ampcode in 2026 you are already falling behind” - this true? crazy if its!
@theisaacmed dont fall into the trap. you just want a new column in your xls file meanwhile the AI wants to sell you a webapp, a database and a lifeinsurance!
@EntireHQ@steipete lets agree to a natural language with limited keywords and some symbols like && or !== , this way we save tokens and have clearer https://t.co/qMzdWaJuAI files!
@unclebobmartin@mattpocockuk In complex domains eg quant finance , our specs often collapse into pseudocode. Code becomes the only unambiguous description.
If we move away from code, what replaces it? first a mess:, later keywords. then we reinvent if/else, loops, UML… until we agree on a master HLL ?
ChatGPT now has full support for MCP Apps.
We worked with the MCP committee to create the MCP Apps spec based on the ChatGPT Apps SDK. Now, any apps that adhere to the spec will also work in ChatGPT.
https://t.co/ybvgXsNX0o