The 20 Generative AI Coding Tools Every Programmer Should Know About
The 20 Generative AI Coding Tools Every Programmer Should Know About
It shouldn’t come as any surprise to learn that today’s generative AI large language models (LLMs) like ChatGPT and Google Gemini are just as fluent in Python, Javascript and C++ as they are in English, Spanish or Arabic!
This has a hugely democratizing effect, as it’s no longer necessary to spend a lengthy amount of time learning to code in order to create simple software tools and applications.
But professional programmers are benefiting too – research suggests the majority of developers are now using the ability of generative AI to fix bugs or suggest alternative methods that can make their code faster and more efficient.
So here’s my rundown of the most powerful and popular tools out there, for everyone from experienced coders to complete novices.
Please note that this list focuses on tools that are specifically designed for coding or coding-related tasks. While ChatGPT, Gemini and other general-purpose chatbots are quite capable of creating code, these specialist tools are built around coding-specific user interfaces, or the ability to integrate into development environments.
Co-Pilot is based on OpenAI’s Codex model, which is itself based on the GPT-3 language model. However, it is integrated into a number of popular integrated development environments (IDE), such as Visual Studio Code, meaning that programmers can easily start using it as part of their existing workflow.
Co-Pilot can generate code from scratch, offer auto-complete suggestions, suggest ways of making existing code more efficient, and aid in debugging. It is fluent in Javascript, Python, C++, C#, PHP and many other languages.
Being based on one of the most powerful language models, it distinguishes itself by being great at interpreting complex and detailed natural language prompts and converting them into usable code. However, as a closed-source model, it has been criticized for its lack of transparency, which can sometimes make it difficult for users to understand why it is making mistakes (as all language models will do from time to time!)
Formerly known as CodeWhisperer, Q Developer is built around Amazon’s proprietary Bedrock model and offers real-time coding suggestions and detailed input and feedback on coding. As part of the AWS ecosystem, it integrates smoothly with Amazon’s other cloud-based tools and platforms, as well as third-party tools and packages. Like Co-pilot, it can read and create code in all of the most commonly used programming languages.
One of its most powerful features is its ability to create autonomous agents, which are essentially unique, personalized AI assistants created to help with specific tasks. They can create and implement testing schedules, automatically refactor code based on best-practice rules, and handle cloud and API integrations.
This generative AI coding tool is mainly designed for use by the enterprise clients who make up IBM’s customer base. It is currently offered as two products. One is Watson X Code Assistant for Z, which is made to help with modernizing old or outdated (legacy) applications. For example, it has models developed specifically for translating applications written in older languages like COBOL and running on mainframe systems into more modern languages like Java. The other is Watson Code Assistant for Red Hat Ansible Lightspeed, which assists in writing code for managing and automating the operation of servers, cloud services and other enterprise IT infrastructure.
Code Assistant is built around a fine-tuned 20-billion parameter version of IBM’s Granite LLM, which is specifically trained on enterprise data that will be relevant for its customer use cases, rather than publicly-available online data.
CodeWP stands out from the other tools here because it is tailored specifically for WordPress coding. WordPress is one of the world’s most popular platforms for building websites and is used by web developers of all levels of expertise. Until now, however, accessing its more advanced functionality and in-depth customization has required getting your hands dirty with some coding. Now CodeWP unlocks these functions for everyone by creating the necessary Javascript, PHP or similar code from natural language prompts. Its language model is fine-tuned on WordPress-specific use cases, and it can even create code for popular plugins such as WooCommerce, used by millions of small businesses to sell directly to customers from their websites.
More Generative AI Coding Tools And Platforms
There are many more generative AI tools that are specifically aimed at helping us write better code or automate routine tasks involved in coding. Here’s the best of the rest:
Augments programming abilities by providing intelligent code completion and optimization.
Boosts software development productivity with a suite of features, including code generation, auto-complete and debugging assistance.
Automated code review and debugging.
Edit, analyze and explain code in real-time designed for enterprise clients.
Meta’s AI coding assistant built on a tuned version of its Llama 2 language model, specializing in creating, analyzing and interpreting code.
Generate code in many languages through a simple web interface.
Generative coding aimed at data scientists and analysts is offered through a Chrome extension.
Automated code review that analyzes code for errors or bugs and suggests fixes.
Open-source AI code generator developed by Carnegie Mellon University researchers.
Automates the creation of programming documentation.
AI-accelerated software development platform with automated test generation.
A cloud-based integrated development environment (IDE) augmented with generative functionality and collaborative features.
Google’s Android Studio platform is used to create applications for its mobile OS, and now comes with built-in coding support powered by Gemini.
AI assistant for software development with strong security credentials.
Not a code writer, but a terminal application that offers contextual autosuggestion, error correction and suggestions on specific commands that a user might want to use.
This tool is designed to help with writing pull requests by automatically recognizing changes between different versions of a source code.
