vladholubiev

Learn how to build dynamic, stateful applications that harness multiple LLMs at once

Does anybody know if there is a similar playground, but for evaluating a single model by comparing different prompts with different temperatures?

I am building this by myself using Streamlit, but was wondering is there a ready solution.

This is excellent! I dreamed of such thing all the time.

Do you mind sharing some technical details on how did you implement data collection, diff, search, aggregation, etc?

Hey there, Vlad, the creator of https://awsnews.info here.

One day I realized that ~30% of AWS product updates updates are some boring stuff like "some service just launched in another region" or "you can now add tags to your appsync api", etc.

So I decided to create a website which filters all the noise out and brings a more pleasant UI than https://aws.amazon.com/new/

Tech Stack:

* Frontend: React App built with Next.js

* Hosting: AWS S3 + AWS CloudFront

* Backend: AWS Lambda + AWS DynamoDB + AWS API Gateway

I haven't used it personally yet, but I want to try out Saasify [1] for my next project.

> Saasify handles all of the SaaS boilerplate, including user accounts, subscription billing, developer docs, and a polished marketing site.

[1] https://saasify.sh/

Since we are on Hacker News, it would be interesting to read something about the tech stack behind the product. After visiting your blog, I suppose you can tell, for example, how just-comments.com provides high availability with the help of AWS Kinesis, DynamoDB, Athena, etc.

By the way, this page is 404: https://just-comments.com/about.html (and some others from ToS)

Otherwise, congrats on the launch and good luck!

I speak Russian and listened to the demos. IMO this AI is trained to be shockingly offensive (especially 2nd call). I feel compassion for the girl answering, she is scared AF. Voice, intonation, and attitude sound like from a debt collector.

P.S. Congrats to the team! This is amazing, just make it more friendly! ;)

I highly recommend this one:

https://www.hello-startup.net

> This book will teach you how to build products, technologies, and teams in a startup environment. It's based on the experiences of the author, Yevgeniy (Jim) Brikman, as well as interviews with programmers from some of the most successful startups of the last decade, including Google, Facebook, LinkedIn, Twitter, GitHub, Stripe, Instagram, AdMob, Pinterest, and many others.

At Shelf [1] we use:

Front-end: React, Redux, Vue.js, Meteor.js. Served from S3 via CDN (AWS Cloudfront)

Back-end: Node.js in AWS Lambda with AWS API Gateway, also Node.js in Docker hosted on AWS ECS, spot instances

Storage: MongoDB (on mLab), Elasticsearch (on Elastic Cloud), S3

Other: GitHub, Circle CI, Terraform, Jest

[1] https://shelf.io/

I don't necessarily read all of them, but ICYMI there is a curated list of engineering blogs - https://github.com/kilimchoi/engineering-blogs

I know this is kinda crazy, but nowadays people run in Lambda Google Chrome and Tensorflow. So maybe this can be a viable idea as well?

Sources on GitHub: https://github.com/vladgolubev/serverless-libreoffice Post on Medium: https://medium.com/@vladholubiev/how-to-run-libreoffice-in-a...