augustinemp
Re:Intro to DeepSeek's open-source week and why it's a big deal
on Thursday March 06, 2025 @01:41PM (#43283605)
I wonder if this openness will inspire other teams to share more of their infra. Imagine how quickly the field could move if more labs took this approach!
Ask HN: Examples of applications that are using o1 style reasoning models
Garry Tan previously described o1 as a huge deal:https://www.youtube.com/watch?v=Lv9bKyQgoHo2 months later, I am wondering what applications have popped up that were not possible with the previous generation of models?
Show HN: LLM reasoning partner for multiple choice questions
https://usmle-study-partner.lovable.app/
I made this for my gf who is studying for US medical exams (USMLE Step 2), which consist of multiple choice questions. The goal of the app is to have an AI study buddy that goes through every part of the question and discusses why or why not it is relevant with you. Here’s how it works:First, it generates a question to answer on a chosen topic.Then once you answer the question, you can...
Re:Radiology specific foundation model released by Harrison.ai
on Tuesday September 10, 2024 @08:04AM (#41499904)
I spoke to radiologist in a customer interview yesterday. They mentioned that they would really like a tool that could zoom on a specific part of an image and explain what is happening. For extra points they would like it to be able to reference literature where similar images were shown.
Re:Ask HN: What are you working on (August 2024)?
on Sunday August 25, 2024 @05:54AM (#41346012)
A companion that you chat with to keep track of your health symptoms. I was inspired by seeing family members go to their doctors appointments and not knowing the important symptoms to report in their short appointments.
Right now, it's Ruth, a companion you can WhatsApp with. She will take your history and record your symptoms, then you can ask her to summarise it before your appointment.
Try out the prototype on WhatsApp: +1 (516) 734-6593
(It's not HIPAA compliant and shares data with LLM providers currently, so best to use a fake patient profile if you're interested)