我的場景下 #DeepSeekV4 Pro 和 #OpenCode 已經接近完美，例如我想 (from scratch) 開發一個 #Rust 版本的RocketChat Client，可以叫它參考官方文檔開發真實測試 #TestSuite 運行並收集數據，再做 Rust Client，它會完全自己測試自己修Bug，例如中間我要求它加入替換用戶名和觸發is typing，這兩個功能官方文檔都沒有寫的，它是直接查看rocketchat的源碼研究，結果亦很完美，未來要維護也有充足的測試
有了 #DFD 和 #TestSuite 作為基礎，再要求AI一邊優化DFD一邊開發代碼，中間用大量的測試作為驗證，可以幾乎零人手參與AI原生進行開發，我使用的是 #DeepSeekV4 和 #OpenCode
Daniel Ward presents 'How to Test Between Microservices Using Contract Testing' July 25th at Nebraska.Code().
https://nebraskacode.amegala.com/
#testing #contracttesting #testsuite #DevOps #Nebraska #csharp #TechConference #opensourcelibrary #software #SoftwareEngineer #softwaredeveloper #programming #coding #qualityassurance #technicaltesting #networkingevent #Tech
'Learn How to Test Between Microservices Using Contract Testing' with Daniel Ward July 25th at Nebraska.Code().
https://nebraskacode.amegala.com/
#LeanTECHniques #microservices #contracttesting #csharp #lincoln #opensource #EndToEndTesting #Nebraska #csharpprogramming #TechnologyConference #softwaredevelopment #testing #testing #SoftwareEngineering #testsuite #opensourcelibrary #opensource
Because I'm a massive nerd, I actually try to read specification documents. As I've ranted ad nauseam before, the current TOTP spec is irresponsibly obsolete. The three major implementations of the spec - Google, Apple, and Yubico - all subtly disagree on how it should be implemented. Every other MFA app has their own idiosyncratic variants. The official RFC is infuriatingly vague. That's no…
Because I'm a massive nerd, I actually try to read specification documents. As I've ranted ad nauseam before, the current TOTP spec is irresponsibly obsolete. The three major implementations of the spec - Google, Apple, and Yubico - all subtly disagree on how it should be implemented. Every other MFA app has their own idiosyncratic variants. The official RFC is infuriatingly vague. That's no…
Over the weekend, I put together an open-source Test Suite for GenAI to help me test the LLMs in the multi-modal GenAI framework. It's a simple tool designed to help developers test and validate Generative AI models across platforms like Ollama, OpenAI, Anthropic, and Amazon Bedrock. Testing is essential in GenAI because it ensures that models generate accurate, reliable, and unbiased outputs—crucial given their complexity and impact.
If you can’t run your #testsuite as part of the #pipeline, then what’s the point? A consistent method of building an unknown quantity for use in #production? That sounds marginal at best.
What isn’t the #documentation for the pipeline telling me? Or what haven’t I seen in it yet? I don’t want to #debug the #testsuite itself. We already know it works from #CLI.
The recording of our Fediverse Developer Network meeting yesterday is now online: https://fedidevs.org/notes/2024-03-07/ and there are meeting notes, too.
Main subject was an early show-and-tell-and-feedback of FediTest, developing a #testsuite for the #Fediverse.
Thanks @andypiper running the meeting and making the recording.
H@R0👨🏻💻
Nebraska.Code
N-gated Hacker News
Hacker News
Mima-sama
Joe Lee
TSM at Work
Fediverse Developer Network