Rami Krispin's Data Science Channel
رفتن به کانال در Telegram
نمایش بیشتر
4 602
مشترکین
+424 ساعت
+27 روز
-1130 روز
در حال بارگیری داده...
کانالهای مشابه
ابر برچسبها
اشارات ورودی و خروجی
---
---
---
---
---
---
جذب مشترکین
ژوئیه '26
ژوئیه '26
+29
در 1 کانالها
ژوئن '26
+36
در 0 کانالها
Get PRO
مه '26
+36
در 1 کانالها
Get PRO
آوریل '26
+34
در 0 کانالها
Get PRO
مارس '26
+31
در 0 کانالها
Get PRO
فوریه '26
+26
در 0 کانالها
Get PRO
ژانویه '26
+30
در 1 کانالها
Get PRO
دسامبر '25
+64
در 0 کانالها
Get PRO
نوامبر '25
+76
در 1 کانالها
Get PRO
اکتبر '25
+140
در 1 کانالها
Get PRO
سپتامبر '25
+156
در 1 کانالها
Get PRO
اوت '25
+120
در 0 کانالها
Get PRO
ژوئیه '25
+133
در 1 کانالها
Get PRO
ژوئن '25
+59
در 1 کانالها
Get PRO
مه '25
+63
در 1 کانالها
Get PRO
آوریل '25
+91
در 1 کانالها
Get PRO
مارس '25
+162
در 0 کانالها
Get PRO
فوریه '25
+54
در 0 کانالها
Get PRO
ژانویه '25
+112
در 1 کانالها
Get PRO
دسامبر '24
+86
در 2 کانالها
Get PRO
نوامبر '24
+68
در 0 کانالها
Get PRO
اکتبر '24
+100
در 0 کانالها
Get PRO
سپتامبر '24
+150
در 0 کانالها
Get PRO
اوت '24
+85
در 1 کانالها
Get PRO
ژوئیه '24
+187
در 3 کانالها
Get PRO
ژوئن '24
+77
در 2 کانالها
Get PRO
مه '24
+213
در 1 کانالها
Get PRO
آوریل '24
+166
در 2 کانالها
Get PRO
مارس '24
+264
در 1 کانالها
Get PRO
فوریه '24
+60
در 1 کانالها
Get PRO
ژانویه '24
+211
در 1 کانالها
Get PRO
دسامبر '23
+159
در 1 کانالها
Get PRO
نوامبر '23
+158
در 3 کانالها
Get PRO
اکتبر '23
+41
در 1 کانالها
Get PRO
سپتامبر '23
+48
در 0 کانالها
Get PRO
اوت '23
+496
در 0 کانالها
Get PRO
ژوئیه '23
+137
در 0 کانالها
Get PRO
ژوئن '23
+42
در 0 کانالها
Get PRO
مه '23
+256
در 0 کانالها
Get PRO
آوریل '23
+41
در 0 کانالها
Get PRO
مارس '23
+6
در 0 کانالها
Get PRO
فوریه '23
+30
در 0 کانالها
Get PRO
ژانویه '23
+35
در 0 کانالها
Get PRO
دسامبر '22
+1 614
در 0 کانالها
| تاریخ | رشد مشترکین | اشارات | کانالها | |
| 26 ژوئیه | +1 | |||
| 25 ژوئیه | +4 | |||
| 24 ژوئیه | +1 | |||
| 23 ژوئیه | 0 | |||
| 22 ژوئیه | +5 | |||
| 21 ژوئیه | +2 | |||
| 20 ژوئیه | 0 | |||
| 19 ژوئیه | 0 | |||
| 18 ژوئیه | 0 | |||
| 17 ژوئیه | 0 | |||
| 16 ژوئیه | 0 | |||
| 15 ژوئیه | 0 | |||
| 14 ژوئیه | +3 | |||
| 13 ژوئیه | +1 | |||
| 12 ژوئیه | 0 | |||
| 11 ژوئیه | +4 | |||
| 10 ژوئیه | 0 | |||
| 09 ژوئیه | 0 | |||
| 08 ژوئیه | 0 | |||
| 07 ژوئیه | +1 | |||
| 06 ژوئیه | 0 | |||
| 05 ژوئیه | +1 | |||
| 04 ژوئیه | +4 | |||
| 03 ژوئیه | +1 | |||
| 02 ژوئیه | +1 | |||
| 01 ژوئیه | 0 |
پستهای کانال
This is a great summary about how to orchestrate AI applications with Graph and tools such as Claude Code 👇🏼
📽️: https://www.youtube.com/watch?v=QRh1a5qvm9U
| 2 | Polars Crash Course - Modern Data Frames in Python 🚀
The following tutorial by NeuralNine provides a one-hour introduction to Polars, moving from installation and core DataFrame operations to expressions, joins, lazy query planning, streaming, GPU parallelism, and plotting. It is useful for Python users who want a structured comparison with Pandas.
📽️: https://www.youtube.com/watch?v=OlsRyy-au0E | 267 |
| 3 | Hermes Agent Fundamentals in 29 Minutes 🚀
The following tutorial by Tina Huang provides a 29-minute introduction to Hermes Agent. It includes a companion resource package and links to related material on local agents and open source AI, making it a useful starting point for exploring the broader ecosystem.
📽️: https://www.youtube.com/watch?v=5_N84t1rUU0 | 323 |
| 4 | Pi Coding Agent Review 🚀
The following tutorial by NeuralNine reviews Pi as a coding agent, compares it with OpenCode, and considers where it fits in local workflows using open-weight models. It is useful for developers evaluating coding-agent options.
📽️: https://www.youtube.com/watch?v=XhliUzOmSa8 | 268 |
| 5 | AI Agents Explained 🚀
This 21-minute tutorial from Tech With Tim gives a practical introduction to AI agents: a language model that can use tools and run in a loop until it finishes a job.
It covers building blocks, the agent landscape, no-code and low-code options, agent harnesses, full-code implementations, and how to choose the right approach.
📽️ https://www.youtube.com/watch?v=ZvDkJsKE80k | 283 |
| 6 | Kimi K3 Coding Tests 🚀
Alejandro AO's 19-minute walkthrough tests Moonshot AI's Kimi K3 against Claude Fable 5, Claude Opus 4.8, and GPT-5.6 Sol.
It covers public coding benchmarks, a real GitHub issue evaluated for quality, cost, time, and token use, reproducing the tests with DuoBench, and using K3 with Pi, Tau, and OpenCode.
📽️: https://www.youtube.com/watch?v=c_XfCbuNRLU | 282 |
| 7 | Issue 97 is out!
This week's agenda:
🔹 Open Source of the Week - skforecast-ai by Javier Escobar Ortiz and Joaquin Amat Rodrigo
🔹 New learning resources - Total Beginners Guide to Local AI on Mac, Run NanoClaw Safely with Docker MicroVM Sandboxes, Python for Beginners with Hands-On Projects, 5 Dockerfile “Best Practices” That Are Actually Wrong, First Look at Kimi K3, AI Agents For Beginners - OpenClaw Case Study, AI Agents Explained
🔹 Book of the week - Practical LLM Evaluation for Production Systems by Ammar Mohanna, PhD, Indrajit Kar, and Zonunfeli Ralte
https://ramikrispin.substack.com/p/the-skforecast-ai-project-practical | 332 |
| 8 |  For my Portuguese-speaking friends: statistical inference, explained in a 12-video playlist 🚀
Prof. Fernanda Maciel's Inferência Estatística playlist is presented in Portuguese. It starts with the core vocabulary of inference and moves into confidence intervals and hypothesis testing, including an applied example in RStudio.
The playlist covers:
✅ Populations, samples, and parameters
✅ Sample statistics
✅ Confidence intervals for proportions
✅ Confidence intervals for means
✅ Hypothesis testing in four steps
✅ Significance levels and p-values
✅ Type I and Type II errors
✅ Hypothesis testing with RStudio
Playlist 📽️: https://www.youtube.com/playlist?list=PLYDRySjwfgb_hcTTO77kK5kiVMxTXk3Nk | 304 |
| 9 | For my Portuguese-speaking friends: statistical inference, explained in a 12-video playlist 🚀
Prof. Fernanda Maciel's Inferência Estatística playlist is presented in Portuguese. It starts with the core vocabulary of inference and moves into confidence intervals and hypothesis testing, including an applied example in RStudio.
The playlist covers:
✅ Populations, samples, and parameters
✅ Sample statistics
✅ Confidence intervals for proportions
✅ Confidence intervals for means
✅ Hypothesis testing in four steps
✅ Significance levels and p-values
✅ Type I and Type II errors
✅ Hypothesis testing with RStudio
Playlist 📽️: https://www.youtube.com/playlist?list=PLYDRySjwfgb_hcTTO77kK5kiVMxTXk3Nk | 1 |
| 10 | First look at Kimi K3 🚀
This 18-minute tutorial from Tonbi's AI Garage tests Moonshot's Kimi K3, a 2.8T-parameter open-weights MoE model, inside Hermes Agent.
It covers the model architecture, long-horizon autonomy, benchmark framing, and three demos: a coding test, a Three.js procedural world, and sub-agent research orchestration.
📽️ https://www.youtube.com/watch?v=Oqk2n3t-CXU | 365 |
| 11 | 5 Dockerfile "Best Practices" That Are Actually Wrong 🚀
The following tutorial by DevOps Toolbox is a short review of Dockerfile advice that may not always improve the final image or application.
It focuses on:
✅ common Dockerfile assumptions
✅ faster image builds
✅ application performance
✅ container infrastructure
✅ deciding when a "best practice" is actually useful
📽️: https://www.youtube.com/watch?v=aZ_y2M2OuEA | 301 |
| 12 | Python for Beginners with Hands-On Projects 🚀
This 8.5-hour freeCodeCamp course by Sunny Dimalu starts with Python and editor setup, then covers variables, strings, operators, conditionals, loops, functions, collections, modules, files, exceptions, OS tools, and sockets.
Projects include calculators, an authentication system, a password cracker, and a MAC address generator.
📽️ https://www.youtube.com/watch?v=oDOw5tB3Udw | 354 |
| 13 | Docker's new tutorial shows how to run NanoClaw inside a MicroVM sandbox on a Mac. It covers the isolation difference from standard containers, sandbox setup, Claude Code installation, NanoClaw configuration, and launching the first agent.
https://www.youtube.com/watch?v=8ZbtZ5TJh0o | 336 |
| 14 |  MiniMax-M3 - a ~428B multimodal model with ~23B active parameters 🚀
MiniMax released M3 with native text, image, and video support, a 1M-token context window, and local deployment options across several inference frameworks.
The efficiency design is the interesting part. M3 uses MiniMax Sparse Attention for long contexts; the project reports 9x faster prefill and 15x faster decoding than M2 at 1M context.
Key features:
✅ ~428B total and ~23B active parameters
✅ Native text, image, and video
✅ 1M-token context window
✅ MiniMax Sparse Attention
✅ Enabled, adaptive, or disabled reasoning
✅ Coding and cowork workloads
✅ Local serving with SGLang, vLLM, and Transformers
✅ KTransformers and Unsloth support
Repo: https://github.com/MiniMax-AI/MiniMax-M3
♻️ Please share if you find it useful | 378 |
| 15 | This is really a great tutorial for running an LLM locally on a Mac. Beyond the Mac setup, great explanation on how to read the model spec on Hugging Face.
https://www.youtube.com/watch?v=ExovHG5FT6s | 391 |
| 16 |  OpenScience - an open-source AI workbench for scientific research 🚀
OpenScience is a browser-based workspace for research agents. You give it a research goal, and it can work through the research loop: read literature, form a hypothesis, write and run code, run experiments, analyze results, and write up what it found.
The project treats scientific research as an agent workspace, not only as a chat interface. It brings together the UI, local server, agent runtime, tools, sessions, and provenance in one open-source project.
Security note: the agent is not sandboxed. The permission system keeps you aware of what the agent is doing, but it is not an isolation boundary, so use a container or VM if you need stronger isolation.
Install with: npm install -g @synsci/openscience
Or run with: npx synsci
Repo: https://github.com/synthetic-sciences/openscience
Docs: https://openscience.sh/docs | 385 |
| 17 | My newsletter is out 👇🏼
🔹 Open Source of the Week - The Tau project
🔹 New learning resources - Self-Hosting Honcho: Free Local Memory for My Hermes Agent, Financial Risk & Performance Metrics in Python
🔹 Book of the week - Autonomous Data Security: Creating a Proactive Enterprise Protection Plan by Priyanka Neelakrishnan, BE, MS, MBA
https://ramikrispin.substack.com/p/tau-autonomous-data-security-issue | 452 |
| 18 | The following tutorial by Nidhi Singh walks through self-hosting Honcho as local memory for a Hermes agent.
It covers:
✅ why self-host agent memory
✅ what Honcho is
✅ Docker + .env setup
✅ running local reasoning + embeddings with Ollama
✅ connecting Honcho to Hermes
✅ debugging connection and embedding-dimension issues
https://www.youtube.com/watch?v=i2KK85h0bYo | 581 |
| 19 | Tau is a readable Python coding-agent harness from Hugging Face 🤗
What I like about it is the separation between the reusable agent core and the terminal coding app.
The architecture has three layers:
✅ tau_ai - provider-neutral model streaming
✅ tau_agent - messages, tools, events, loop, harness, and sessions
✅ tau_coding - CLI, Textual TUI, file/shell tools, project instructions, skills, prompts, and on-disk sessions
A useful detail: the core emits typed events instead of rendering UI directly, so the same harness can drive print mode, the TUI, or another frontend.
Install:
uv tool install tau-ai
Repo: https://github.com/huggingface/tau
Docs: https://twotimespi.dev/
Licence: MIT | 565 |
| 20 |  All the talks from the DuckCon conference 🦆 are now available:
https://www.youtube.com/playlist?list=PLWksJtu3SS6Y
#data #duckdb #dataengineering #datascience | 491 |
