If you are tired as me of working with native path and fs libraries in NodeJS, you might like my
www.npmjs.com/package/@sy... NPM package.

A library to finally turn verbose JS filesystem code into elegant one-liners

LLMs have helped me build dozens of libraries I've always wanted in Javascript

My favorite one is
www.npmjs.com/package/@sy...

The concept is simple, what if you could write clean template strings in JS ?

I've been using it for clean error formating, LLM messages, and more

If you're building LLM based products, check out napkin.ai

Blending pre-made templates with LLM formatting may seem like a straightforward idea but they've executed it really well with a clean UX

And if you need to fill a presentation with corporate BS, just ask it to generate the text as well 😅

While illustrating the many disciplines connected to AI and LLMs I realized that I don't have a term for UX Designers / PMs focused on AI

Really shows how under-explored UX for AI is

I agree it's hard, but we could be doing so much more than chats

I already have a good vision on how to achieve combination 1+2+3 and have started building a PoC

My dream would be to incorporate a real canvas as well like the one in my UI Sketcher plugin for VSCode/Replit

@vercel.com's AI SDK is amazing to work with

But I don't get why I had to implement a middleware to add the last message from the prompt to the result if its role is "assistant"

Should be a default behavior

Anyway, if you need it it's "includeAssistantMessage" in @synstack/llm/middleware

4️⃣ Prompt chaining

Because prompt scripting is mostly async, I disliked writing await and Promise.all to whenever working on message arrays

So, the message() param can be a Promise<Array<Message>> or an Array<Promise<Message>>, making it simpler and cleaner to build out your prompt and chain agents

3️⃣ Template strings

I hate the base JS template string because it breaks indentation
So I wrote a separate library to fix that.

🔶 Trims text and removes extra indentation
🔶 Auto-resolves promises
🔶 Arrays to separate lines
🔶 Propagates nested indentation

github.com/pAIrprogio/...

For v1, I wrote my own abstraction layer

While I love the chainable + immutable + string-templates + type-safety combo, writing my own adapters was too time consuming

So after using @mattpocock.com Evalite, I decided to keep my approach but switch the engine to Vercel's sdk as a peer dependency

Got my first gaslighting from #Devin 🎉

It introduced a regression by overstepping its objective and removing some linter ignore comment

Then claimed that the issue was pre-existing

But "learned" from it

I feel like that's not going to be the only occurence

What I liked on this session:

- It handled a failing pipeline and identified some wip code that was failing the tests
- It asked me how I wanted to deal with it
- It built knowledge from my answer
- It's a one shot task, writing the prompt myself would have been bothersome

I've gone through this with Reforge, my own tool for AI assisted migrations

V1 used a custom templating engine and most edge case required either new development or an ugly "hack" of the engine

So for V2, I went with TS scripts with some custom syntax highlighting

Yet the hardest challenge in JS is handling the amazing but verbose async context

And do we want "await" to appear in our prompts and mess up with the readability ?

Of course we don't, so we handle parrallel promise resolving for you

And of course, because joining arrays manually is a drag, we do it for you

No more having to specify ".join("\n")"

Let's keep that prompt ✨beautiful✨

Indentation is very important to preserve instructions hierarchy in prompts. But it's a mess to handle with traditional template strings

So, when you interpolate synstack/text with a multiline string, the starting indentation will be replicated accross every line

When prompting, every token counts. But JS template strings requires content to be de-dented, which I find harder to read and maintain

So with synstack/text, trimming and de-denting is done for you.

No more messy codebase

But that's far from the only feature...

So glad for the opportunity to talk with @fab-ber.bsky.social about this amazing journey of turning 10 months of work into 4

Thanks to @reactnativelondon.co.uk for hosting us and to @delphinebugner.bsky.social for this amazing sketch