Viewing profile — chenglong-hn

no yet! that's a good direction!

Mostly for expressiveness and reliability & cost trade-off. Flint has advantage of being an intermediate language that allow agents to generate good-looking stuff without additiona…

Both, it's a combination of latency / reliability / expressivenss. I think when models are more powerful, we will also more naturally want to create more expressive stuff.

This is awesome!

We struggled the same, especially when deploying an agent to end users, they are even more frustrated with any mistakes from agents than us!

This is a formal language that compiler can take! The natural language and clarify loop helps the LLM to generate Flint spec that can render.

We made an intentional decision to make it non-compositional for now, and we are using chart-type based approach to handle more expressive charts. (e.g., CDF charts) But! We have a…

Thanks a lot for the suggestion, I do think we will need to work on accessibility support in Flint, this is a perfect place to centrally handle accessibility issues. I added an iss…

We will link it on github and the project page! https://github.com/microsoft/flint-chart

We find an additional dimension where previous language doesn't explicitly use (semantic type) as the key part of the specification, which turned out to be the root reason of why c…

That's the best option for now. But it can also be frustrating to ask AI to do small edits on D3 just to fix some idioms (like switching order etc) and they kept messing up with ot…

I find it lacks some easy way to do alignments or grouping, which makes editing frustrating. Could be a good language re-design opportunity.

great point, I feel both the low-level foundation language and the high-level abstraction should evolve together, just like how human programming language evolves (e.g., frontend f…

Yes! correctness, expressiveness, reliability and cost (token + time)

the design here has some constructs (i.e., semantic types) that AI can use better than human at generation time; and then the generated spec can be easy for user to edit since ther…

Also guidelines sent to the agent may or may not get ignored if they are just part of the context :(

more like, instead of watching the AI agent 20 minutes to create a polished chart with 100k tokens, Flint lets you get there in 20 second with 200 tokens with a 95% polished chart.…

It's often comes with missing keys, use wrong value type (e.g., list over dictionaries). Mostly a small model issue and open source models, they don't follow instructions on the st…

You are absolutely right. But note that we are actually on the same point here. This is exactly why this is an intermediate language designed to get 95% stuff right easily (for exp…

A little bit more than that! Here shows why a higher-level abstraction can be helpful for chart generation: https://github.com/microsoft/flint-chart#features

It's a white box! We built a demo here: https://microsoft.github.io/flint-chart/#/documentation/char... We use a fun elastic algorithm to decide dimension etc within the developer'…

also user interaction afterwards -- if can be frustrating if the only way the user can interact with the chart is to chat with the agent again (simple spec allows easy UI interacti…

JSON schema can do it to some degree, I think?

Seems a bit more like https://mermaid.js.org use cases!

I felt conflicted as well, json is portable and easy to parse / validate and edit. But many models do still struggle. There are some stuff from functional programming might be wort…