Very good

Generative AI is improving at such a fast pace that I doubt whether after sometime, we would simply have to put in proper prompts and the AI itself will be able to code another AI. Just some days back I used the Bard AI to code some simple lines on how I could integrate my external AI dubbing / text-to-speech software with my smartphone at home that my family uses. I used teh codes and didn't work as well as I wanted it too, but nonetheless, it was impressive 👍🏽

Thanks for sharing Anton's impactful use of ChatGPT in software development! It's amazing to see how AI techniques like voiceover or text to speech is aiding individuals like Anton in overcoming unique challenges.

❓ How are you incorporating large-language models (LLMs) such as ChatGPT or GitHub Copilot into your development workflow? Share your favorite models and tips! Whether it's MLOps, custom projects, or enhancing test-driven development (TDD), let's chat about the exciting intersection of human creativity and AI in coding. Explore more about AI voice over in development.

I leverage Copilot for fast prototyping for some python and js stuff, using precise prompts to generate modular code and tests, boosting my TDD workflow by ~30%. what's your tool to boost performance?

I'd like to highlight the importance of context and how context engineering can significantly boost the output quality of any generative AI model being used.

Context engineering is about shaping not only your prompts but also the supporting information you feed to the model, such as examples, background data, prior chat history, or even “system”/instruction blocks. Carefully curating context enables:

• More relevant, on-topic, and accurate model responses
• Improved handling of multi-turn interactions or larger workflows
• Better performance when chaining or orchestrating multiple models/tools
• More consistent outputs

From my experience, spending time on engineering the right context window (what to include/exclude, what reference materials to attach, etc.) has a bigger impact on output quality than tweaking the core prompt itself, especially for more complex, production-grade use cases.

For the most part, you will be good having the following files in your workspace when building anything with the help of AI:

• architecture.md - This defines your project file structure, data flow, state management etc.
• tasks.md - granular step by step task breakdown for MVP development and testing.
• design.md - design principles/guide you want the model to follow while designing anything.

As for my favorite models right now, I’m especially impressed by:

• GPT‑5 (incredible versatility, reasoning and instruction following)
• Claude Sonnet 4 (overall just amazing)
• GLM 4.5 (super cheap and actually good performance)

I use them through Cursor and Kilo Code if you're curious.

Hope this helps!

The ReadME Project by GitHub celebrates the heart of open source—its people and their ideas. It’s a home for the stories, insights, and innovations shaping the software you rely on every day. From in-depth articles to practical guides, The ReadME Project shines a light on the creators driving the future of open source.

Subscribe to get fresh perspectives, inspiring stories, and proven best practices from the global open source community—delivered straight to your inbox. EPDS Bihar

The ReadME Project by UP Ration Card GitHub honors the essence of open source—its people and their powerful ideas. It’s a dedicated space for the stories, lessons, and breakthroughs that define the software shaping our world. Through thoughtful features, deep dives, and hands-on guides, The ReadME Project spotlights the innovators and communities building the future of open source collaboration.

IDME KPM adalah platform digital berpusat Kementerian Pendidikan Malaysia yang memudahkan pengurusan identiti dan pengesahan data dalam sektor pendidikan. Sistem ini menyediakan Single Sign-On (SSO) untuk mengakses pelbagai aplikasi pendidikan seperti DELIMa dan eRPH. Ibu bapa boleh mendaftarkan anak dan mengurus penempatan sekolah, guru boleh mengurus pengajaran dan rekod pelajar, manakala pentadbir sekolah boleh menyelia maklumat sekolah, guru, dan pelajar. IDME KPM memastikan keselamatan data dan pengurusan pendidikan yang cekap dalam ekosistem digital Malaysia.