Viewing profile — hellovai

(on of the creators of BAML here) yep! exactly! that workaround we've found works quite well, but the problem is that its not sufficient to just retry in the case of failed schema …

if you haven't tried the research -> plan -> implementation approach here, you are missing out on how good LLMs are. it completely changed my perspective. the key part was really j…

Have you tried techniques that don’t require modifying the LLM and the sampling strategy for structure outputs? For example, schema aligned passing, where you build error tolerance…

if you share your prompt with me on promptfiddle.com i can play around with it and see how i can make it better!

its a bit more nuanced than applicative lifting. parts of of SAP is that, but there's also supporting strings that don't have quotation marks, supporting recursive types, supportin…

appreciate you tyring it. the reason it dropped the day was due to your type system not being understood by the LLM you're using. the model replied with { "Text": "coffee liqueur",…

have you tried schema-aligned parsing yet? the idea is that instead of using JSON.parse, we create a custom Type.parse for each type you define. so if you want a: class Job { compa…

really cool to see BAML on here :) 100% align on so much of what you've said here. its really about treating LLMs as functions.

yea! even deepseek. Calling an external function / tool calling is really just a data extraction problem. say you have a tool: def calculator(expr: str) -> float then the model jus…

hey all! Vaibhav and Ethan here. We asked a silly question, "what if you attempted to look into every commit of every repo?" and over the holidays ended up building something kind …

Hey HN, We made an LLM Playground where you can use jinja to render prompts and extract structured data. For example, receipts, invoices, audio snippets, etc. It uses the BAML prog…

i 100% percent agree. people get so caught up on trying to do everything 90% right with AI, but they forget there's a reason most websites offer at least 2 9's of uptime.

We have some preliminary data with llama3.1 and we find that the smaller model gets to around 70% with BAML (+20% from base), but we'll update this dashboard with llama3.1 by end o…

Take a look at BAML (boundaryml.com) Its a different take that leverages a DSL to make prompting cleaner and fixes a few other ergonomic issues along the way. you can try it online…

nothing specific, but you can try our prompt / datamodel out on https://www.promptfiddle.com or if you're open to share your prompt / data model with, I can send over my best guess…

we recently added dynamic type support with this snippet! (docs coming soon!) Python: https://github.com/BoundaryML/baml/blob/413fdf12a0c8c1ebb75c... Typescript: https://github.com…

oh thats really interesting, how often do you get errors like that? fyi, we actually fix those specific errors in our parser :)

The main drawback is really when you attempt to do more advanced prompting techniques like chain-of-thought or reasoning. forcing those parts to be json, can be hard and unnecessar…

;) https://www.promptfiddle.com/structured-summary-66myE (sorry bad syntax highlighting when including baml code in baml code) { author: "Sam Lijin" key_points: [ "Structured outpu…

that's a great question, there's three main benefits: 1. seeing the full prompt, even though that python code feels leaner, somehow you need to convert it to a prompt. a library wi…