- Big Data News Weekly
- Posts
- 🤖How to Speed-Up Training of Language Models
🤖How to Speed-Up Training of Language Models
🦾Plus: 🌐 ChatGPT Gets Ready for Ads
In partnership with
Hey folks! Let’s get into Big Data and AI craziness…
In today's edition: What's Shaping the Future of Data?
🧱Exploring the Architecture of Large Language Models
🕸️Large-Scale Interactive Training with Monarch
💬Agents Should Be More Opinionated
💻How good engineers write bad code at big companies
🤖 Amazon Connect adds autonomous AI agents
🧩 Silicon Valley leans on free Chinese AI
💡 AI Tutorial:Create Instagram product shots with Nano Banana Pro
🤖 AI Tools and Data Tools to checkout
Language model training is slow, even when your model is not very large. This is because you need to train the model with a large dataset and there is a large vocabulary. Therefore, it needs many training steps for the model to converge. However, there are some techniques known to speed up the training process. In this article, you will learn about them. In particular, you will learn about: Using optimizers Using learning rate schedulers, Other techniques for better convergence or reduced memory consumption.
Your competitors are already automating. Here's the data.
Retail and ecommerce teams using AI for customer service are resolving 40-60% more tickets without more staff, cutting cost-per-ticket by 30%+, and handling seasonal spikes 3x faster.
But here's what separates winners from everyone else: they started with the data, not the hype.
Gladly handles the predictable volume, FAQs, routing, returns, order status, while your team focuses on customers who need a human touch. The result? Better experiences. Lower costs. Real competitive advantage. Ready to see what's possible for your business?
Artificial Intelligence (AI) is no longer a distant notion; it is very much a current transformational force. There is a hint of AI in almost everything, from your Netflix account to real-time translation of languages. Right at the core of a number of these intelligent systems is a powerful tool: The Large Language Model (LLM).
Monarch (Meta's distributed actor framework) combines the power of large-scale training with the familiarity and ease of interactive development. This template showcases multi-node training using Monarch (Meta's distributed actor framework) with TorchTitan (PyTorch's large-scale LLM training library) on Lightning AI infrastructure. You'll learn how to set up, execute, debug, and manage distributed training workflows across multiple GPU nodes.
The best agent products are the most opinionated. The goal in agent products is to give users a delightful experience. A good baseline for agents is that everything works reliably without tweaking too many settings. Good product design is the result of creators distilling their vision into an intuitive interface that just works.
For engineers working on self-contained technical projects, the only explanation for bad code is incompetence. Other engineers operate more like plumbers or electricians, working on projects with awkward or surprising parts that are relatively new to them. In these cases, bad code is inevitable, but as long as the overall system works well enough, the project is a success.
think-cell integrates seamlessly with Microsoft Office on both Windows and Mac. Secure architecture ensures deployment is fast and reliable, so your teams can start maximizing their productivity immediately. New to think-cell? Try it in a free 30-day trial and accelerate your enterprise-wide efficiency. No charges. No credit card needed.
👨💻 Data Tools, Libraries
Coresignal’s Free Webinar: How to Evaluate External Data Quality with Confidence. Join Coresignal’s free, vendor-neutral webinar on December 2, where Data Analyst Egidijus Griska will share practical tools and metrics you need to be confident when assessing B2B data.
kit (GitHub Repo)
kit is a toolkit for codebase mapping, symbol extraction, code search, and building LLM-powered developer tools, agents, and workflows. It can build things like code reviewers, code generators, and IDEs.
Styleframe (Website)
Styleframe's powerful TypeScript CSS API helps developers compose design systems in minutes.
AI News:
Amazon Web Services announced 29 agentic AI capabilities for Amazon Connect at its re:Invent conference, featuring fully autonomous AI agents that can handle complex customer service requests across voice and chat channels without human intervention, while seamlessly transitioning to human representatives when needed.
Shoppers are adding to cart for the holidays
Peak streaming time continues after Black Friday on Roku, with the weekend after Thanksgiving and the weeks leading up to Christmas seeing record hours of viewing. Roku Ads Manager makes it simple to launch last-minute campaigns targeting viewers who are ready to shop during the holidays. Use first-party audience insights, segment by demographics, and advertise next to the premium ad-supported content your customers are streaming this holiday season.
Read the guide to get your CTV campaign live in time for the holiday rush.
OpenAI will test ads inside ChatGPT, starting with search. Code references in the latest Android beta reveal formats like carousels, product-style listings and new "bazaar" content types. The rollout comes as ChatGPT usage reaches staggering levels, nearing 800 million weekly users and billions of monthly visits.
Aristotle, an AI system built by Harmonic, just independently solved a 30-year-old Erdős problem, marking what researchers are calling the first real step into the “vibe proving” era of mathematics. Aristotle solved a version of Erdős Problem #124, which has been open since the 1990s, in six hours, and then formally verified the proof in Lean in a minute.
Chinese open models have surged, giving US startups fast, cheap tools that rival closed systems. Developers pick these models for flexibility, strong community support and lower costs. This shift exposes how few high-end open models the US has, sparking new government and industry moves to catch up.
The new Tesla Ride program brings supervised Full Self-Driving demos and Grok AI-guided experiences to consumers. Participants will be allowed to sit in the driver's seat while a Tesla Advisor rides in the front as co-pilot. The sessions are capped at 45 minutes each, and participants are required to have a valid driver's license and insurance. The program will run at some locations until the end of December.
Q4 is the perfect window to turn this year’s numbers into a clear, actionable forecast aligned with your goals. Set your business up for a stronger 2026 with BELAY’s new guide.
AI Tutorial
🖼️ Create Instagram product shots with Nano Banana Pro
In this tutorial, you’ll learn how to use Nano Banana Pro to generate a full 9-image Instagram feed from just one inspiration photo, turning your product shots into cohesive, high-quality visuals for social media campaigns.
Step-by-step:
Go to Gemini → Tools → Create Images, ensure Pro mode is enabled, and upload an inspiration image that reflects your desired style or aesthetic
Upload your product image, describe it, then prompt with: “Create a 9-image Instagram feed for this product with varied angles, people, and environments”
Click Submit to generate your 9-image grid. Review results and, if needed, ask Nano Banana to regenerate or isolate specific shots
Download your favorite visuals and post them directly to Instagram, TikTok, or your brand’s storefront for an instant, consistent feed
What 100K+ Engineers Read to Stay Ahead
Your GitHub stars won't save you if you're behind on tech trends.
That's why over 100K engineers read The Code to spot what's coming next.
Get curated tech news, tools, and insights twice a week
Learn about emerging trends you can leverage at work in just 10 mins
Become the engineer who always knows what's next
🔥Top AI tools to increase productivity:
Marblism generates a fully-functional web application from a single prompt:
Clipwing A tool for cutting long videos into dozens of short clips
aiPDF is an innovative, multi-modal tool designed to work with a wide array of inputs, including ebooks, web articles, YouTube videos, podcasts.
podcast.ai, a podcast that is entirely generated by artificial intelligence
Clipwing A tool for cutting long videos into dozens of short clips
aiPDF is an innovative, multi-modal tool designed to work with a wide array of inputs, including ebooks, web articles, YouTube videos, podcasts.
Verk- Hire AI employees to add more firepower to your team, who work 24/7 to do sales, be your personal assistant, do graphic designing and more
Codetoflow enables you to understand the code in simple terms using a flowchart which enables you to understand the details
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
👀 little avengers
Recommended reading:
SPONSOR US
Get your product in front of Big Data & AI enthusiasts
Our newsletter is read by thousands of tech professionals, investors, engineers, managers, and business owners around the world.
Interested in Sponsoring the Big Data News Weekly Newsletter?Get in touch today
What did you think of today's email?Your feedback helps me create better emails for you! |