Rami Krispin’s Data Science Channel

Rami Krispin’s Data Science Channel

5/16/2025, 3:55:39 AM

Hyperparameter Tuning in Python 👇🏼 The following tutorial by Code with Josh focuses on hyperparameter tuning machine learning models with Python. This includes random search and grid search tuning approaches using SciKit. https://www.youtube.com/watch?v=JJpVCJGg7Z8

Rami Krispin’s Data Science Channel

5/16/2025, 1:24:32 AM

I had the pleasure of presenting today at the ODSC East 2025 conference about querying data with LLMs. Creating an AI agent (which is a fancy name for three Python functions) is easier than it sounds. At its core, all you need: ➡️ A function that creates a prompt based on a user input ➡️ A function that sends the prompt to an LLM and parses the answer into SQL code, and ➡️ A function that picks up the query and sends it to a database to get the data All the workshop materials are available in the following repo: https://github.com/RamiKrispin/osdc-2025-llm-workshop

Rami Krispin’s Data Science Channel

5/17/2025, 3:07:58 PM

My weekly newsletter is out! This week's agenda: ✅ Open Source of the Week - The DataMap project by Steven Ge ✅ New learning resources - New tutorials for RAG, linear regression with Python, ML foundation, stats for data scientists, hyperparameter tuning ✅ Book of the week - Scaling Up with R and Apache Arrow by Nic Crane, Jonathan Keane, and Neal Richardson https://ramikrispin.substack.com/p/the-datamap-project-scaling-up-with

Rami Krispin’s Data Science Channel

5/16/2025, 12:09:41 PM

If you are looking for some weekend learning, my LinkedIn Learning course - 𝑫𝒂𝒕𝒂 𝑷𝒊𝒑𝒆𝒍𝒊𝒏𝒆 𝑨𝒖𝒕𝒐𝒎𝒂𝒕𝒊𝒐𝒏 𝒘𝒊𝒕𝒉 𝑮𝒊𝒕𝑯𝒖𝒃 𝑨𝒄𝒕𝒊𝒐𝒏𝒔 𝑼𝒔𝒊𝒏𝒈 𝑹 𝒂𝒏𝒅 𝑷𝒚𝒕𝒉𝒐𝒏 is free for a limited time. https://ramikrispin.substack.com/p/limited-time-github-actions-course #rstats #python #github #data

Rami Krispin’s Data Science Channel

5/14/2025, 7:09:56 AM

Faster Data Pipelines development with MCP and DuckDB 🚀 This looks super awesome - developing data pipelines faster using DuckDB and an MCP server 👇🏼 https://www.youtube.com/watch?v=yG1mv8ZRxcU

Rami Krispin’s Data Science Channel

5/16/2025, 12:24:53 PM

Apologies, in case the link is not leading you to the free option, please use the following link: https://www.linkedin.com/posts/rami-krispin_happy-friday-if-you-are-looking-for-some-activity-7329117038877069312-dTT9?utm_source=share&utm_medium=member_desktop&rcm=ACoAAAtOwDsBZIC3sJSnivdI1ISUXm_8nd-oEf0

Rami Krispin’s Data Science Channel

5/19/2025, 3:17:20 PM

Langchain Agents 🚀 The following playlist by TheAILearner provides an introduction to different use cases on AI agents using the Langchain framework. This includes using Langchain to build: ✅ SQL agent ✅ Custom agent ✅ Custom agent with memory ✅ Dynamic few-shot prompting https://www.youtube.com/playlist?list=PLkEeXQbdBATjEM9F7UqDU9unwQvm2TBMm

Rami Krispin’s Data Science Channel

5/18/2025, 10:13:16 PM

Linear algebra, along with calculus and probability, is the foundation of data science and AI 🚀. Whether you are a student or looking to brush up on your linear algebra skills, the MIT Linear Algebra course by Prof. Gilbert Strang is one of the BEST resources ❤️. This full-semester course is an entry-level Linear Algebra course, and it covers the following topics: ✅ Core matrices operation - elimination with matrices, multiplication and inverse matrices, transpose, permutations, etc. ✅ Column and null space ✅ Solving Ax matrices ✅ Projection matrices and least squares 😎 ✅ Differential equations ✅ Linear transformations and their matrices 📽️: https://www.youtube.com/playlist?list=PL49CF3715CB9EF31D

Rami Krispin’s Data Science Channel

5/17/2025, 10:05:12 PM

Fine-Tuning Local Models with LoRA in Python LoRA (Low-Rank Adaptation) is a technique for fine-tuning large language models by injecting trainable low-rank matrices into each model layer, allowing adaptation with significantly fewer parameters. This makes training more efficient, memory-friendly (and mainly cheaper) while preserving the original model weights. This one-hour tutorial, by NeuralNine, focuses on the theoretical and practical approach for fine-tuning LLMs with LoRA using Python, and it covers: ✅ Theory & Mathematics ✅ Fine-Tuning on Math Problems ✅ Evaluation of Math Problems ✅ Fine-Tuning on Custom Data ✅ Evaluation of Custom Data https://www.youtube.com/watch?v=XDOSVh9jJiA

Rami Krispin’s Data Science Channel

5/15/2025, 2:31:23 AM

Statistics for Data Science 🚀 The third and last article from the Data Hustle newsletter 🗞️ focuses on core statistical concepts for data science. This article covers foundational topics such as the statistical relationship between variables, correlation, margin of error, statistical power, time series, etc. Thanks to Sai Kumar Bysani and Vaishali Macwan for the great summary! https://thedatahustle.substack.com/p/advanced-statistics-for-smarter-business