[Dev Catch Up #19] - Claude 3 Opus, Meta AI, Stable Diffusion 3, and more.
Bringing devs up to speed on the latest dev news from the trends including, a bunch of exciting developments and articles.
Welcome to the 19th edition of DevShorts, Dev Catch Up!
I write about developer stories and open source, partly from my work and experience interacting with people all over the globe.
Some recent issues from Dev Catch up:
Join 1000+ developers to hear stories from Open source and technology.
Must Read
The GPT-4 model from OpenAI is considered as the leader in the chatbot section because of its sheer dominance in terms of LLM competence in different factors along with market share. It has been a leader in the Chatbot Arena, a leaderboard run by the Large Model System Organization which is a research organization dedicated to open models. This organization’s site allows visitors to rate outputs from using various models and this allows the site to calculate the best models in aggregate. But GPT-4 is not the frontrunner in the leaderboard because of Anthropic’s latest model “Claude 3 Opus”. Claude’s rise is a great leap towards the advancement of different AI models and definitely will hit a pause to GPT-4’s popularity marathon, although it is a year old. The article from Arstechnica gives detailed information on the output of Claude’s performance, how it beats GPT-4 in the game, and the future of advanced LLM models.
The battle to claim the crown in the AI battle is getting interesting day by day. After Google and Microsoft, another tech giant Meta has now entered the battle ring with OpenAI’s ChatGPT. Recently, the company unveiled its Meta AI assistant that is being integrated into search boxes within its applications. The assistant, alongside being accessible by a standalone website, will start appearing directly in the Facebook feed and users can chat with it directly with the help of Meta’s messaging apps. It is built with Meta’s highly talked Llama 3 model which is a top performer on key benchmarks and outperforms competing models of its class. Apart from pulling out real-time search results from both Google and Bing, it can also generate images. This article from The Verge gives you great insight on the Assistant and its features.
With the current condition in the world of AI, the battle to create the best LLM model is already thriving with a lot of innovation. Generative AI models are the talk of the town these days and Stability AI adds one more reason to keep the conversation going with their Stable Diffusion 3 Generative AI model. As of today, you can use the model as an API on Stability AI developer platform. Making the model production-ready along with the API usage was a key focus for the company. With the implementation of a number of safeguards to prevent the model misuse, the company is planning to release an open model in the coming days. It has also shown a beta preview of the Stable assistant, a chatbot built on the company’s texts and image generation technology. Learn more about the new model and the assistant from this article published by Venturebeat, where a detailed analysis on both the products and the future of the AI space has been laid out.
Now, we will head over to some of the news and articles that will be at a place of interest for developers and the tech community out there.
Good to know
The framework that has revolutionized CSS over the recent years is Tailwind. Recently v4.0.0 of Tailwind got released and it came with a bucket full of improvements and additions. The upcoming version will get a number of deprecations as well. With the new high-performant Oxide engine in Tailwind CSS 4.0, building the CSS website becomes 10 times faster. 4.0 comes with backward compatibility meaning that it provides support to a bunch of important configurations, functions, prefix, and modes. The 4.0 version comes with a removal of deprecated utilities. In version 4.0, the tainwindcss package separates the postCSS plugin and CLI, unavailability of any border color by default, and the ring utility is set to 1 px by default. More information on the update is available in this article published from the official Tailwind team.
Apache Kafka is one of the most popular tools in the software world and one of the first choices for big organizations to store and exchange data. But it has a problem with querying the data easily. So, SQL is the best bet to query data while using Kafka for data communication, storage, and production. Kafka produces the data before being moved to other applications but that data is almost not accessible for non-developers and product owners. Here is a brilliant article from TheNewStack that explains the querying of those barely accessible data with the help of SQL.
In today’s world, Redis is one of the most popular in-memory datastore that is used as a database, cache, and message broker. The high popularity and demand for this database is mainly because of its speed, simplicity, and versatility. Now, there is another open-source alternative to Redis that is backed by AWS, Oracle, and Google. Valkey, the high performance data structure, suddenly came to the limelight when Redis Labs, the creators of Redis imposed restrictive licensing. Valley serves primarily key-value workloads and furthermore, supports a wide range of native structures and an extensible plugin system for adding new data structures and access patterns. Learn more about Valkey from this well explained article from Cyberkendra, where detailed information on its licensing and developer adaptability is given.
Celebrating open-source projects in every edition has been a regular lately and with this edition, there is no exception as well. The open-source project with a bunch of stars and gaining a bit of attention is “otel-profiling-agent”. It is developed by Elastic, a pretty big open-source organization. The agent is a whole-system cross-language-profiler for Linux via eBPF. With a very minimal CPU and memory overhead, it implements the experiential OTel profiling signal. The agent supports C/C++ executables, Arm64 support for all languages except NodeJS, profiling of system, mixed stackraces, native codes, HLLs, and much more. Check it out from its GitHub repository here and leave a star if you like it.
Lastly, we will take a look at some of the trending scoops that hold a special mention for the community.
Notable FYIs
AI is shaping the future of technology in every way possible and as days are passing by, it is eminent that research on AI will become the next big thing. AI research is crucial for both businesses or organizations and individuals. This latest podcast from Latent Space talks about how AI research can be supervised with AI UX and how Elicit, an AI research assistant company is helping with supervised AI research.
Observability is one of the prime technologies when it comes to examining the nits and bits of your application and it became revolutionized with the use of OpenTelemetry. This video titled Shift Into an Observability Mindset with OpenTelemetry from Dan Gomez Blanco’s session from Observability day Europe 2024 gives a deep inside knowledge on embracing OpenTelemetry into your application monitoring.
Architectural patterns are reusable solutions for the common architectural design problems that are encountered during software development. Here is a short article from ByteByteGo that explains the 9 architectural patterns for data and communication flow.
The payments gateway is a tech wonder that has evolved over time with the revolutionization of ecommerce. To understand this technology, you need to understand the ecosystem around it and how it connects everything with major payment networks. Here is a short article with visualization from ByteByteGo where they have explained the payment ecosystem and the steps associated with it in layman terms.
A few days ago KubeCon 2024 EU ended and the sessions are as good as it can get with loads of information on shaping the future of Kubernetes and cloud-native ecosystem. All of those sessions are now available in Youtube and you can have a look at all of them from this playlist from the official CNCF Youtube channel.
That’s it from us with this edition. We hope you are going away with a ton of new information. Lastly, share this newsletter with your colleagues and pals if you find it valuable and a subscription to the newsletter will be awesome if you are reading for the first time.