Is it open source, can you share? Have been pondering using source generators to achieve a similar effect and have the LLM build the source generator ‘framework’ that code would derive from rather than just generating the code itself, you create the code generation framework.
What really interests me most is the sharding and the possibility of using this for multitenancy - is the hooks / plugin architecture sufficient so you can run a small sidecar to add shards or tenants to the TOML file dynamically? Would be a game changer.
I’m sure the use of duckdb may seem weird for normal developers, but for data people it really is game-changing, especially for data scientists or business analysts.
I literally didn't read any comment that spoke to the context to which PG was speaking - Oxford Union.
Any sane, rational person would offer different advice to people they see as "future prime ministers and billionaires" or at the very least people influential in future policy decisions, than they would to society in general.
Someone needs to build Qt’s successor, probably with more beginner-friendly declarative semantics (akin to HCL or Cue) and probably with syntax closest to YAML or Python (based on learning curve, beginner readability etc).
The backend will probably have to be something written in Zig (likely) or Nim (capable, less likely) and will probably have to leverage OpenGL/Metal, WebGL and WASM.
Obviously a massive undertaking, which is why I think the industry has not reached consensus that this is what needs to happen. The less ideal options we have now often gets the job done.
This should be the top comment. The thing about growth of businesses overall, is that they want outsourced capacity (that’s what employees or contractors are) and that dynamic doesn’t go away because of AI, because like the comment mentioned, it’s not reliable enough in the sense that it can accept high-context vague instructions and ‘figure it out’ like an enterprise developer can.