- Big Data News Weekly
- Posts
- 🤖 Make Your Own LLM Agent in R
🤖 Make Your Own LLM Agent in R
🦾Plus: 💸 Binance Was Hit With a Record $6.3Bn Crypto Heist
In partnership with
Hey folks! Let’s get into Big Data and AI craziness…
In today's edition: Here is roundup of Mistral Large 4, Sentinel Values in Python, Binance hack, 👇
🧠Next-Concept Modeling: Fast, Parallel Coarse-to-Fine Text Generation
💻Mistral releases open coding and cyberdefense model
🐍Much Ado About None: Sentinel Values in Python
🛰️ Google's Space Data Center Satellite Wrote its First Haiku
💡 AI Tutorial:How to Get Access to Claude Duet
🤖AI Tools and Data Tools to checkout
It’s really easy to make your own agent. You should do it once, just to understand how this technology works. LLM agents are becoming pervasive, with Meta’s Muse perhaps the first consumer grade agent. They are only going to get more common. If you know a bit of R you can program your own, and it makes the whole thing feel a lot less like magic.
Nobody's coming to approve your idea
People assume that real businesses get built by other people with more resources. It's how great ideas get stuck in notes apps and conversations.
But you don't need six months and a developer's salary to validate your concept. You just need a weekend. Describe what you want to exist and Lovable builds it: a real site, a real product, ready to accept real payments from real customers. Not a mockup to show people. An actual thing they can use.
You don't need anyone's sign-off to get started. Take $5 off your first month.
Discount of $5 applies to first month of any paid plan. You need to sign up via this webpage in order to claim the discount. The discount applies to new and returning users without an active paid plan. Terms apply.
DevOps introduces a culture of collaboration, automation, and continuous improvement. When applied to data science—often referred to as MLOps or DataOps—DevOps practices help streamline the entire lifecycle of data-driven applications.
Diffusion language models (DLMs) offer parallel sampling for faster generation than autoregressive models, an increasingly important advantage as inference becomes a dominant computational workload. However, DLMs degrade quickly with fewer sampling steps, process the full sequence at every step, and require many passes over the dataset to reach good quality.
🚀 Dive into the World of Machine Learning System Design! 🌐🔍
Ready to uncover the secrets behind how industry giants like Netflix, Airbnb, and Doordash leverage machine learning for unparalleled success? 🤖✨
📘 Download 'ML and LLM System Design: 800 Generative AI Case Studies & 45 ML Platforms': Unlock a treasure trove of insights with 800 real-world case studies from 100+ companies, delving into practical ML use cases and lessons learned in designing ML systems.
Mistral Large 4 is our largest and most capable model to date: a 1 trillion-parameter natively multimodal Mixture-of-Experts model with 49 billion active parameters per token (52 billion including embeddings and output layers), trained from scratch on 3,800 NVIDIA Grace Blackwell GPUs in Mistral's own datacenters in Europe.
None is probably the most overloaded value in Python. We use it as a Swiss Army knife to express that something is missing, unset, that an error occurred, or simply as an actual value. In my talk about sentinel values, which I presented this year at PyConDE/PyData Berlin 2026, EuroPython 2026 and, most recently, the Swiss Python Summit 2026, I take a closer look at why this ambiguity bites us and what to use instead…
Start a 30-day Snowflake trial and run real workloads: build data pipelines, work with open Iceberg tables, and deploy AI on your data, all on a fully managed platform that runs across clouds. Build reliable pipelines: create continuous data pipelines in the language of your choice.
👨💻 Data Tools, Libraries
VectorAI DB: Do you need a vector database? Find out in 30 seconds. Before you budget for a vector database, check whether you need one. Answer a few questions about your data, latency, filtering and deployment.
Try Snowflake free for 30 days with $400 in credits and run your analytics, pipelines and AI on one platform. Load your data, run SQL at any scale, and build pipelines and ML models without managing infrastructure.
Headroom (GitHub Repo)
Headroom is a context optimization layer for LLM applications that cuts costs by 50% to 90% without losing accuracy.
AI News:
Binance just had the kind of Friday that makes a security team polish its résumés. Hackers working for North Korea have been blamed for stealing at least $6.3 billion in crypto after compromising an engineer's laptop, the largest known crypto theft in history. About half was reportedly frozen as it moved across blockchains. Binance has suspended deposits and withdrawals and insists user funds are safe.
Free email without sacrificing your privacy
Gmail tracks you. Proton doesn’t. Get private email that puts your data — and your privacy — first.
Google's Suncatcher satellite is online, and its first act in orbit was poetry. The prototype carries four TPUs and ran DeepMind's Gemma 4 to write a haiku about space, then reported back on conditions around it. The catch is power: it can only switch one chip on for about 20 minutes at a time. Even orbital AI has a battery problem.
Waymo is finally getting a Toyota to play with. Under a partnership announced Friday, Toyota will build 2027 bZ4X electric vehicles fitted with Waymo's self-driving system at its Takaoka factory in Japan. Testing starts in Japan next year, with the U.K. and other right-hand-drive markets to follow. It lands 18 months after a preliminary deal vague enough to count as "let's see what happens."
The AI shopping revolution is still waiting on its first big order at DoorDash, which says it has received only "several hundred" orders from third-party agents. For scale, it completes about 3 million deliveries a day. CEO Tony Xu told analysts you never want to fight gravity, but you also don't want to be delusional. Most customers still just use the app.
Somebody put a $60 million AI supercomputer on eBay. The listing was 100% real, and it even had a Make Offer button. A hardware reseller in Minnesota listed 300 racks of computers, and they said they came out of a big cloud provider’s data center. Each rack stands about as tall as a refrigerator and is packed with Nvidia’s A100 chips, the workhorse behind a lot of AI a few years back.
TLDR AI is a free daily email curated by Anthropic and ex-Google engineers: the new tools worth trying and the research worth understanding, summarized for 1.1M+ readers.
AI Tutorial
How to Get Access to Claude Duet
I just got access to Claude Duet, Anthropic's new AI assistant that can work for you even when you're offline. Claude Duet connects to your apps, manages tasks, and keeps working in the background. You can give it instructions through text or voice, and it will notify you when the work is finished or needs your approval. Here is how it works
Visit the official Claude Duet website. Click Join the waitlist.
Enter your name and email address associated with your Claude account.
Enter your company name.
Choose your phone type (iPhone or Android).
Submit your application and check your email for an invitation.
Once invited, open the invitation on your phone, follow the installation instructions, and sign in with your Claude account.
Connect the apps you want Claude Duet to work with, such as Gmail, Google Calendar, Google Drive, or Slack.
Jobs:
This week's Data & Analytics jobs: I have updated my curated list with fresh openings. Browse and apply in one click: jobs
Machine Learning Engineer – Hybrid – Boston
Lead Machine Learning Engineer – Texas
🔥Top AI tools to increase productivity:
Best Face Swap is an online tool that lets you replace faces in digital media using various AI models.
Framia Pro is an all-in-one AI platform for creating and editing images and videos
xLeadForge is an X outbound workspace for people who need to find and contact the right audience
Ito is an AI-powered quality assurance platform built for modern software teams
Creatii AI helps creators, marketers, and brands generate polished AI videos, images
dOCR is one API for documents and the web
View our database of all the best AI tools for your needs: aitoolsup.com
Have cool resources to share? Submit AI tool
A.I. Generated Image of the Day
👀 Happy Halloween
Recommended reading:
SPONSOR US:
Get your product in front of Big Data & AI enthusiasts
😮 The Marketing Channel You've Probably Slept On:
Haven't tried newsletter sponsorships yet? You are missing out on a HUGE ROI+ customer acquisition channel. I know because dozens of advertisers keep coming back for more... Run a test campaign with us and see for yourself 👉 Get in touch today.
What did you think of today's email?Your feedback helps me create better emails for you! |