[Dev Catch Up #22] - Ajax LLM, ChatGPT Memory, Angular 18, and much more.
[Dev Catch Up #22] - Ajax LLM, ChatGPT Memory, Angular 18, and much 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 22nd 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
Improvements in the field of AI and ML are seen on a daily basis and new features on products are constantly popping up from the market leaders in the technology. Recently, OpenAI introduced the Memory feature that allows ChatGPT to have a more permanent storage of queries, prompts, and other customization. Memory in ChatGPT works in two ways. It lets you tell the assistant to remember certain details and also learn from conversations. Once remembered, it includes your preference without the need for a reminder. Learn more about this feature from this article published by The Verge where more details on the feature have been explained in detail.
In the recent days, there are ongoing discussions on the various AI-related updates and enhancements that will come to Apple tools and applications with the release of iOS 18. Like all other tech giants, Apple doesn't want to lag behind in the AI race and has been working on its own Ajax LLM. Safari can bring a Text Summarization feature which is one of the top ones in the works at the moment. Siri will also receive the same type of feature which will generate responses that will consist of relaying content of the messages in a simplified way. The LLM will generate basic responses entirely on-device. More advanced replies or summaries will require server-side processing. Learn more about AI updates in iOS 18 from this article published by Appleinsider where they have talked about the new features in detail with interactive graphics.
In this world of Artificial Intelligence, everything is becoming autonomous with complex tasks becoming simpler and easing human life. The same thing if happens with programming will make a developer’s life easy by letting the machine do the complex tasks. With the rise of Large Language Models or LLMs and Generative AI reeling into the modern world, the innovation in the field of code generation through simple spoken language prompts has taken pace. Generating texts and extracting information is easy with LLMs. Dive deep into the world of LLM with this article published from FireworksAI where they explained how code generation through LLMs is playing an important role in building new age experience for developers and improving their productivity drastically.
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
One of the popular frontend frameworks Angular has released its new version, v18, and it came with a lot of new features and enhancements. The new version comes with the enhancement of evolving change detection with experimental support for zoneless to achieve interoperability. A series of improvements can be observed with the increased stability of Material 3, deferrable views, and built-in control flow. More developer enhancements are implemented with the improvements in server-side rendering, such as introduction of il8n hydration support, better debugging, hydration support in Angular Material, and event replay which is powered by the same library as used by Google Search. Finally, Angular devs will also get a new documentation page which will contain in-depth tutorials, interactive playgrounds, and improved search experience powered by Algolia. Apart from all these, there are much more improvements and additions which you will be able to know in detail from this well explained article from the Angular team in which all the updates and improvements are explained in depth with interactive demos and code-snippets.
With AI and LLM centering most of the discussions in the tech world, it is important to understand the limitations that come with this technological advancement. Modern applications will use LLMs but the extensive time and cost required to train these models on large datasets make these models almost impossible to retrain regularly. These challenges bring lack of updates in the model with the latest data which lead to potential inaccuracies when queried about unfamiliar topics. This phenomenon is called hallucination and to overcome these challenges, several techniques are used among which, the most famous is Retrieval Augmented Generation or RAG. It is used because of its efficiency and performance. Usama Jamil from MyScale has written an informative tutorial which is published on Newstack on how to design a complete advanced RAG system that can be used in production environments.
With every technology being assisted with AI in recent days, it is a no-brainer to assume that the future of Cloud-native will also get an AI push. Kubernetes adoption is hard and most organizations face a lot of challenges during its adoption. To ease this issue, Nutanix is aggressively expanding its Cloud-native support with AI. The goal is to provide AI-driven solutions for operations built on Kubernetes. Nutanix has started making advancements on that and has presented a demo on how their AI-powered automation can do in practice for Kubernetes infrastructure. This article from the Newstack gives perspective on Nutanix vision on providing an AI push to end Kubernetes adoption issues.
Every week, we tend to find one of the buzzing open-source projects or tools that has garnered some attention among developers. The project that has raked up quite a handful of stars is Contrast. Developed by Edgeless Systems, it enables developers to deploy and manage confidential containers on Kubernetes at scale. With data encryption at the focus, it prevents access from the infrastructure layer. It is compatible with managed Kubernetes and integrates with your existing Kubernetes workflow. Contrast helps you increase the security of your containers, move sensitive workloads, shielding of code and data from own cluster admins, simplifying regulatory compliance, etc. Check out Contrast from their GitHub page here and leave a star to support them.
Lastly, we will take a look at some of the trending scoops that hold a special mention for the community.
Notable FYIs
The hype forming around a particular technology like AI is not new as the same happened with the emergence of Kubernetes and Cloud computing technology. Over the past decade, both these technologies have emerged and are still hot on the plate with the deliverables of scalability, efficiency, and operational flexibility. This type of swift scale of technological adoption brings the challenge of tackling configuration technological debt which reduces developer productivity and increases security risks. The same can happen with AI with the revolution going around it. This short informative article from Newstack brings a detailed perspective on what we can learn from Kubernetes and the cloud that can steer the AI revolution without incurring technological debt.
JavaScript is one of the top programming language choices for developers more than ever in 2024 and mastering this language is essential for a good fullstack developer experience for modern web development. This article from the Newstack discusses 5 cutting edge JavaScript techniques that will help developers master the language and show new and innovative ways to build dynamic web applications full of interactivity and performance.
Histograms in Prometheus are not new as it has been around the observability platform for more than 10 years but understanding them is not an easy task for most. Here is a YouTube video from the one and only Prometheus expert Julius Volz explaining all about histograms in Prometheus in a nutshell.
Latency has a critical role in AI applications as it contributes with a significant edge in terms of good product experience. LLMs are often expensive to operate and due to their high latency, it is hard to scale generative AI applications in production. Here is an informative article published by the Newstack that talks about the importance of “Cost of Ownership” and latency in AI applications.
In the cloud-native ecosystem, gRPC plays a crucial role as a remote procedure framework that can run in any framework. The gRPC conference 2024 is the largest conference where members of the gRPC community gather around and discuss the developments in the field of gRPC. The Call for Papers is open and you can contribute your thoughts by submitting a session. If it interests you, submit your talk from the official gRPC Conf page from the Linux Foundation here.
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.