New article going round about #ai copilots and #llms #llm being found not useful.
When’s the last time you trusted a study on real world software teams productivity? Would you trust a study that says “agile found to have not productivity improvements”? When there are about 800x ways you can do “agile” and 2000 ways to measure “productivity”.
I encourage people to look at people who have the results to show they are credible and check out how they use ai tools (looking at @simon).
1/
Personally, here’s what I built in about 1h at a coffee shop yesterday using cursor and various #llm #llms:
- autoreload websocket thingamajig with docs and demo https://github.com/go-go-golems/clay/tree/main/pkg/autoreload
- (minimal for now) svg generating dsl, with full spec, examples, unit tests: https://github.com/go-go-golems/go-go-labs/blob/main/pkg/svg/svg-dsl.md
- live web server that renders svgs and autoreloads if a svg gets updated: https://github.com/go-go-golems/go-go-labs/blob/main/cmd/apps/svg/main.go
I also fixed a few issues in other packages.
2/
Personally, here’s what I built in about 1h at a coffee shop yesterday using cursor and various #llm #llms:
- autoreload websocket thingamajig with docs and demo https://github.com/go-go-golems/clay/tree/main/pkg/autoreload
- (minimal for now) svg generating dsl, with full spec, examples, unit tests: https://github.com/go-go-golems/go-go-labs/blob/main/pkg/svg/svg-dsl.md
- live web server that renders svgs and autoreloads if a svg gets updated: https://github.com/go-go-golems/go-go-...