Mikyo

Two recent quotes changed the way I think about developing code: 🎤 “One of the risks I see is fossilization — of the libraries and tools we use.” – Laurie Voss from @llama_index (linkedin.com/posts/auth0_de…) 🎤 "...they're computers but they are humanlike [...] there's people…

📈 @ArizePhoenix now has project dashboards! In the latest release @arizeai Phoenix comes with a dedicated project dashboard with: 📈 Trace latency and errors 📈 Latency Quantiles 📈 Annotation Scores Timeseries 📈 Cost over Time by token type 📊 Top Models by Cost 📊 Token…

Wild.

This is a great post by @sanjanayed and aligns well with what @HamelHusain and @sh_reya pitch in their evals course as well. You don't want to outsource your annotations. It makes a lot of sense to use tools that let you build your own annotation tools (using @v0, @lovable_dev…

The biggest thing preventing @cursor_ai and Claude Code from being magical for me is that they forget to run code formatters like prettier and ruff. I always have to intervene at the end.

TIL about jsonc - how did it take me so long. Now can we get package.jsonc ?

So many enterprises have been asking for LLM tracing support in Java and Kotlin. Excited to make this happen via @ArizePhoenix #oss

The deer in headlights look in the eyes of interview candidates that are addicted to cursor.

Vibe coding? Try vibe understanding the generated code before you tab tab tab or press accept first

AI for code editing is not perfect and often times frustrating. But as an OSS maintainer, there are things that are starting to come into view. What tools like @cursor_ai can help you do is shift your role - rather than the code producer, you become the compiler. The AI takes…

I created an annotation version of @beirmug 's presentation on IR Evals for RAG Nandan argues that we should consider additional retrieval metrics beyond the classics (MRR, etc) b/c retrieval goals for RAG can sometimes be very different hamel.dev/notes/llm/rag/…

PSA: don’t use products that were sold to you, use the ones that you genuinely love and enjoy

Monorepo wins every time.

Aravind Srinivas: “have the intellectual humility” to know what you’re good and not good at. youtu.be/2jOnoTEk-xA?si… I also love that he triages bugs. 🐛

Stayed here a month ago. Definitely was taken aback when I stepped inside.

It’s happening! #Golang github.com/modelcontextpr…

Tracing and telemetry traditionally has been an operational requirement, not a development one. But I've found that with AI applications this fundamentally changes. Take LLM-as-a-judge for example. You might pick an off-the-shelf eval library and trust that it works. But this…

This is what they call "an agent."

I’ve wondered this myself. How do you even evaluate this? @cursor_ai add ways to do pairwise evals or something

Using the tool I helped build to build the tool I'm building. This is the way. #Oss