[Dev Catch Up #39] - Logic-of-Thought Prompting, Stripe toolkit with LLM agents, Cline AI, Common Corpus dataset, Essential tsconfig.json options, 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 39th 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
Evolution of LLMs have come a long way but their performance with complex logical reasoning tasks are unsatisfactory. To address this issue, Logic-of-Thought prompting is introduced which employs propositional logic to generate expanded logical information from input context and utilizes the generated information as an additional augmentation for the input prompts. This enhances the capability of reasoning. Read more about it from here.
The development of AI agents is on the rise and Stripe has risen to this occasion to show that their toolkit can be integrated to the AI agents. This will enhance the functionality of those agents by enabling access to financial services and tools. This article shows how to integrate the Stripe toolkit with an LLM agentic workflow.
Optimizing videos in an application leads to a smooth user experience. To achieve this, a multimedia processing tool called FFmpeg comes to the rescue and along with Node.Js, video streaming can be improved leading to a better user experience. This tutorial gives you details about it.
Bun supports V8 APIs without using V8 JavaScript engines. V8 engines are used by Node.Js but Bun uses the C++ APIs that interact with the JavaScriptCore. This topic comes with a two-part article that describes how Bun pulled this thing off.
After this, shouting out a top open-source project is a delight and here it is:
OSS Highlight of the Week
This week, we focus our attention on Cline, which is an AI assistant that can use the CLI and editor of a developer. Powered by Claude 3.5 Sonnet’s agentic coding capabilities, it can handle complex software development tasks in consequent steps. Learn more about it from its GitHub page and leave a star to support it.
Hope you are enjoying this edition of our newsletter so far! Support us by giving a free follow to our LinkedIn and X pages.
Your support is highly appreciated!
Good to know
To support the development of AI in an open manner, the training of large language models needs to be done with non-copyrighted data. Addressing this, Pleais released Common Corpus, the largest open multilingual dataset for training LLMs. Here is an article containing more information on the dataset and check out the dataset from its Huggingface page.
Cosmos DB is the high availability NoSQL relational database from Microsoft and is one of the most popular ones in the market. Together with Azure, Microsoft’s cloud computing platform, it becomes more powerful in terms of productivity. This is beautifully explained from this article that guides you when to use Cosmos DB with practical examples.
The tsconfig.json file in TypeScript is very useful because it acts as a tool that can manage the behavior and security of the code. This article points out the essential tsconfig.json options that one should use.
Achieving offline support while using TanStack Query is a challenge when maintaining full control over the data layer. This article dive deep into this issue with a solution that supports offline mode while using TanStack Query.
Vision Language Models or VLMs struggle to perform systematic and structured reasoning while handling complex visual question-answering tasks. This problem gets addressed with the launch of Llava-o1, a novel VLM that is designed to conduct autonomous multistage reasoning. Learn more about this model from here.
Lastly, we will take a look at some of the trending scoops that hold a special mention for the community.
Notable FYIs
Stories on LLM development are more on the focus with the latest developments happening all over AI and tech. One such story involves Lindy.ai and how they evolved beyond the LLM-pilled era, their customer acquisitions, and a discussion on why agents need rails. This podcast tells you all about it.
Progressive Web Apps or PWAs are a great way to enhance the UI of an application. These apps can be accessed offline after installation. Here is an article on dynamic generation of PWA app icons with a Next.js app router.
I think we are all somehow familiar with the term “Data Migration” that essentially indicates digital transferring of data from one place to another. This article deep dive into the concept.
Any software solution with zero downtime is great to have because of the high availability it provides to the user. But this type of solution becomes very useful during internal migration for an organization. Here is an article (broken down into 5 parts) that discusses designing a zero downtime migration solution with data consistency.
The hype with AI-agents is real and they are sitting in the driver’s seat in carrying the AI developmental growth. The agent software ecosystem is evolving significantly and you can learn more about it from this guide that groups the different AI-agent stacks for common technological use-cases.
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.