galgia
galgia on Hacker News
Posted by
galgia on Thursday January 15, 2026 @08:26AM
from the show-hn dept.
https://puzzle-crush.vercel.app/
Old school aesthetic and new learning concepts.The whole mastery plan consists of 18k puzzles. Split by each tactical concept, each piece, each square.Example:
Unique chess squares: 64
Topic: M1
Unique pieces: 6
Puzzles: approx. 6*64The coverage of squares is 85%, as some moves are mathematically impossible, and for some exotic positions, I did not find any quality puzzles.Have fun and master...
Every time I start a new project I have to collect the data and guide clients through the first few weeks before I get some decent results to show them. This is why I created a libary to quickly build classic data science pipelines with LLMs, which you can use to demo any pipeline and even use it in production for non-critical use cases.
Good point! LLMs are best when you are starting from point 0.
LLMs are not the most efficient way to solve the problem, but they can solve it.
Exactly, scale after you need to.
Yes, LLMs are not always the best option, they are an option. Sometimes requirements of the project are such that they are also the best option.
There is one browser that uses price matching example that is impossible to do without a full-blown data science team right now: https://github.com/Pravko-Solutions/FlashLearn/tree/main/exa...
+ I assumed that most people will ctrl+a -> ctrl+c -> ChatGPT -> ctrl+v
I belive that LLMs will become better and better in the near future and pipelines will replace classic approaches with LLM-enriched pipelines will drastically simplify the ETL flows.
If your problem is compute, you are already optimizing. This is here for all the steps before you start thinking latency-compute. Not all use cases are made equal.
I see it as a gray area - long term there will be a need for both and you will have just one tool to choose from when presented with time-budget-quality constraints.
You are right! This is here to be used when your resources do not allow you to build full-blow solutions. Yes, I used LLMs to help create examples from my existing code, but they are based on things I have put in production when the client's resources were limited and wanted to move from point 0 to test out the potential of LLMs on their data.
Posted by
galgia on Sunday February 09, 2025 @06:39AM
from the github.com dept.
Posted by
galgia on Sunday February 09, 2025 @05:53AM
from the show-hn dept.
https://github.com/Pravko-Solutions/FlashLearn
I have built many AI agents, and their development always felt weird because all frameworks were mangling together everything from data stores to data processing, which felt super weird to me as a software engineer.I built Flashlearn which is completely JSON driven(JSON in, JSON, out) and it only supports interacting with LLMs everything else you do exactly like you would in normal software...
Posted by
galgia on Friday February 07, 2025 @05:16AM
from the flashlearn.tech dept.
Posted by
galgia on Thursday February 06, 2025 @11:25AM
from the github.com dept.
Posted by
galgia on Thursday February 06, 2025 @05:49AM
from the github.com dept.
Posted by
galgia on Wednesday February 05, 2025 @05:10AM
from the github.com dept.
Thank you! It took a while :)
Posted by
galgia on Tuesday February 04, 2025 @08:36AM
from the show-hn dept.
https://github.com/Pravko-Solutions/FlashLearn
Every time I wanted to use LLMs in my existing pipelines the integration was very bloated, complex, and too slow. This is why I created a lightweight library that works just like scikit-learn, the flow generally follows a pipeline-like structure where you “fit” (learn) a skill from sample data or an instruction set, then “predict” (apply the skill) to new data, returning structured...
Posted by
galgia on Monday February 03, 2025 @10:10AM
from the github.com dept.
Every time I wanted to use LLMs in my existing pipelines the integration was very bloated, complex, and too slow. This is why I created a lightweight library that works just like scikit-learn, the flow generally follows a pipeline-like structure where you “fit” (learn) a skill from sample data or an instruction set, then “predict” (apply the skill) to new data, returning structured results.
High-Level Concept Flow
Your Data --> Load Skill / Learn Skill --> Create Tasks --> Run Tasks --> Structured Results --> Downstream Steps
Each step results can be turned to JSON and reused later or elsewhere.
Installation:
pip install flashlearn
Posted by
galgia on Monday February 03, 2025 @05:07AM
from the github.com dept.
Posted by
galgia on Saturday February 01, 2025 @11:39AM
from the show-hn dept.
https://github.com/Pravko-Solutions/FlashLearn
Working with LLMs in existing pipelines can often be bloated, complex, and slow. That's why I created FlashLearn, a streamlined library that mirrors the user experience of scikit-learn. It follows a pipeline-like structure allowing you to "fit" (learn) skills from sample data or instructions, and "predict" (apply) these skills to new data, returning structured results.High-Level Concept...
Minimal Perplexity clone.
What it does:
- Input user query
- Generate 1 to n google queries to find the answer
- Use this queries to query Google
- Generate response with links to sources
You can retrieve from any source line this and improve the quality of answers for multi-hop questions from your users.
Have fun!
Posted by
galgia on Thursday January 30, 2025 @10:09AM
from the github.com dept.
When you want AI to perform tasks independently, it's important to be able to control it and understand its actions. Additionally, you need a way to organize the results and manage the inputs efficiently. FlashLearn simplifies this process by using JSON definitions to structure everything clearly. This structured approach ensures that you can easily define tasks, process results, and handle inputs, making AI management more straightforward.
Yes, you can! The responses are always structured, you will have to read the files yourself (as img/text/audio or mixed) and then you can feed them to the FlashLearn agent to complete tasks with them.
Yes, you can find them on GitHub in the folder /examples. There you will find more samples and flows on how to use it. The examples are elementary, but they should give you an idea of how to build one.
Posted by
galgia on Thursday January 30, 2025 @03:47AM
from the show-hn dept.
https://github.com/Pravko-Solutions/FlashLearn
I have built many AI agents, and all frameworks felt so bloated, slow, and unpredictable. Therefore, I hacked together a minimal library that works with JSON/dict/kwargs definitions for each step, allowing you a simpler way to define reproducible agents. It supports concurrency for up to 1000 calls/min, giving you speed and predictability in your workflows.Install pip install flashlearn
Input...