Ali loves life

Previously, I shared Chuan Shi’s excellent lecture notes on factor investing. Here are 4 more from the same series, covering ML, factor timing, and alternative data. Great reading if you’re building factor strategies. - Machine learning in factor investing papers.ssrn.com/sol3/papers.cf… - Factor timing and factor allocation papers.ssrn.com/sol3/papers.cf… - Alternative data papers.ssrn.com/sol3/papers.cf… - Behavioral finance and factor investing papers.ssrn.com/sol3/papers.cf…

@JA_Olaoye Totally agree. I’ve learned more fixing broken pipelines at 2AM than from any Kaggle dataset. That’s where data science stops being theory and starts creating business value.

80% of people learning Data Science today are wasting their time. Yeah… I said it. Because most of them are: • training models on clean Kaggle datasets • building portfolios that never touch real systems • chasing “AI” without ever fixing a single data pipeline Meanwhile, in the real world… Companies are struggling with: • data that arrives late • numbers that don’t match across reports • pipelines that break at 2AM • executives who don’t trust their own dashboards But you want to build a recommendation model? On what exactly? Fantasy data? Here’s the truth nobody wants to say: If your data engineering is weak, your data science is useless. Not “less effective” - useless. Because in production: • bad data kills models • inconsistent data destroys trust • slow pipelines make insights irrelevant The real power right now is not in fancy models… It’s in owning the flow of data. Data Engineers are not “support roles.” They are the ones quietly deciding whether your entire data strategy succeeds or fails. But sure… keep learning another regression algorithm. I’m watching.

@nlw I get this completely. Starting has never been easier, but staying relevant takes real strategy and focus. That balance between speed and sustainability defines the modern startup game.

This is the hardest and easiest its ever been to build a startup. No barriers to entry but also no moats is a wild combination that no one has fully recalibrated to yet.

4 AI Certifications for those interested 🔥🔥 1. Intro to AI Ethics kaggle.com/learn/intro-to… 2. AI matters open.edu/openlearn/educ… 3. Elements of AI course.elementsofai.com 4. Ethics of AI ethics-of-ai.mooc.fi

@mdancho84 Great breakdown. Clear grounding and auditing steps make RAG practical for building reliable AI workflows in enterprise settings.

2. Steps (retrieve → ground → reason → act → audit) Retrieve: expand the question, search indexes/DBs, rank hits. Ground: select/snippet the most relevant passages/tables. Reason: synthesize an answer/plan using only grounded context. Act: return a response, call tools (e.g., SQL, web, email), or generate artifacts. Audit: check citations, factuality, policy; log traces for replay.

80% of AI beginners are confused by RAG Agents. Let's fix that (in under 2 minutes): a thread🧵

Claude is literally teaching me maths right now and i actually understand it?? like it just showed me WHY a positive medical test doesn't mean you're sick (Bayes theorem) with a live interactive dot grid and i could drag sliders to see it change in real time normal distribution, central limit theorem, full interactive bell curves all in one chat this is how school should have worked

Data manipulation has never been so simple as with dplyr in R. To help you master this topic, I created an online course titled "Data Manipulation in R Using dplyr & the tidyverse." Check out the course: statisticsglobe.com/online-course-… In the course, you'll learn how to handle your data like an expert. You'll get access to: - 21 video lectures. - An interactive group chat for questions and exchange. - Practical exercises and sample projects. - Comprehensive R scripts and further resources. I've released a sample module to preview what the course offers. In the sample module, you'll learn to handle rows of a data set using dplyr functions such as filter(), slice(), and arrange(). Learn more: statisticsglobe.com/online-course-… #tidyverse #RStats #R #DataAnalytics #Data

Axelar is proud to join the @Mastercard Crypto Partner Program, connecting cross-chain infrastructure with trusted global payments. On-chain payments are going mainstream.

🔥Breaking: @llama_index announced LiteParse for local document parsing. No GPU, No waiting. 100% Open source

These 4 DevOps & AI guides are FREE today ⚡ No fluff. Just real, practical content: - DevOps Roadmap 2026 → 5-phase framework - AWS ECS & EKS → 25 hands-on labs - AI Agent Builder’s Handbook - GitHub Actions → visual guide (v1.0) Learn → Build → Ship faster. Grab them before they’re gone 👇 Looking for more free guides like this? Share it with your friends and support the journey 🚀 t3pacademy.gumroad.com

The wide range of ggplot2 extensions for data visualization in R is truly impressive. Even better, these extensions usually work seamlessly together, making it easy to enhance your plots. Below is an example of an animated ggplot2 plot created using the gganimate and ggblend extensions. Just brilliant! The visualization shown below is taken from the ggblend package website. You can also find the code there: mjskay.github.io/ggblend/ If you’re interested in mastering data visualization in R with ggplot2 and its extensions, you might want to explore my online course on "Data Visualization in R Using ggplot2 & Friends"! Check out this link for more details: statisticsglobe.com/online-course-… #RStudio #tidyverse #coding #Data #database #ggplot2 #Rpackage #DataVisualization #RStats #programming

The future of AI isn’t faster answers. It’s better thinking. And MiroThinker-1.7 is a step in that direction. Explore: Hugging Face: huggingface.co/collections/mi… GitHub: github.com/MiroMindAI/Mir… #AIResearch #ResearchAgents #AIReasoning #ArtificialIntelligence #MiroMind

The Agentic #AI Maturity Gap: Combining Orchestration, Observability & Auditability buff.ly/0QUpbDW @Hackernoon #GenAI Cc @Fabriziobustama @gezgintrk @VallesMaxime @FrRonconi @bzarkout @crimson_crypto @rvp @domingonarvaez1

🧮 Principia: Training LLMs to Reason over Mathematical Objects 📐 We release: - PrincipiaBench, a new eval for *mathematical objects* (not just numerical values or MCQ) - Principia Collection: training data that improves reasoning across the board. For models to help with scientific and mathematical work, you need to train on such data & test whether they can derive things like equations, sets, matrices, intervals, and piecewise functions. We show that this ends up improving the overall reasoning ability of your model for all tasks. Read more in the blog post: facebookresearch.github.io/RAM/blogs/prin…

Data Science courses with Certificates (FREE) ❯ Python cs50.harvard.edu/python/ ❯ SQL kaggle.com/learn/advanced… ❯ Tableau openclassrooms.com/courses/587360… ❯ Data Cleaning kaggle.com/learn/data-cle… ❯ Data Analysis freecodecamp.org/learn/data-ana… ❯ Mathematics & Statistics matlabacademy.mathworks.com ❯ Probability mygreatlearning.com/academy/learn-… ❯ Deep Learning kaggle.com/learn/intro-to…

@quantscience_ Dynamic risk models reshape the efficient frontier. I see Bayesian methods making it practical, and UAE fintech spaces applying this precision globally.

There's a curve in finance that most investors get wrong. It's called the efficient frontier. Markowitz defined it in 1952. Most investors still don't understand what it means in practice. Here's what they get wrong:

I recently found this Python roadmap, and it’s an excellent resource for anyone looking to strengthen their Python programming skills or advance their career as a Python Engineer. It lays out the key areas to focus on, from foundational concepts to advanced topics, making it ideal for learners at any stage. Here are some of the essential areas to explore: ✔️ Core Python: Start with the basics, including algorithms, file handling, exception handling, and data structures like lists, sets, and dictionaries. Dive deeper into advanced topics such as iterators, generators, and lambda functions. ✔️ Software Development Tools: Learn essential tools for improving code quality, testing, and debugging. Familiarize yourself with package managers like PyPI and Poetry, along with version control and virtual environments for efficient project management. ✔️ Web Development: Explore frameworks like Django and Flask for building web applications. Gain expertise in working with databases, REST APIs, GraphQL, and deploying web services using tools like Docker. ✔️ Data Processing and Handling: Master data serialization, working with relational and non-relational databases, and cloud-based solutions like MySQL, DynamoDB, and Elasticsearch. ✔️ Advanced Python: Deepen your knowledge with topics like design patterns, memory management, garbage collection, and metaclasses. Learn to optimize your code using C extensions for high-performance applications. I found this roadmap on the AIGENTS website, and what makes it particularly useful is its interactive design. Each section is clickable, offering AI-powered explanations and curated learning resources to help you dive deeper into each topic. It’s a comprehensive and effective guide to mastering Python. Learn more: aigents.co/learn/roadmaps… #DataAnalytics #RStats #Data #Rpackage #DataScientist #datascienceeducation

NEW: @ApexGlobalGroup commits $100B in tokenized assets to @trex_network, a compliance layer for RWAs The network is built using @0xPolygon CDK and connects to @Agglayer

Polygon CDK is becoming defacto stack for Institutions to build their compliant chains 3.5T dollar Apex group unit Tokeny and Polygon came together T-Rex chain, the protocol behind billions of dollars in RWA with commitment of 100B worth of RWAs by end of 2026. Multiple Big T players are coming onto the CDK. Stay Tuned. Trillions! Literally.

@nlw I agree completely. When coding skills reach that level, broadening into analysis and business tasks is a natural next step. It’s focused growth, not distraction.

Saw a couple people say this shows a lack of focus. I think it reflects the discovery that once your app can code really well, it can do a whole bunch of other things really well too.