Every company is a data company.
Now, of course, every company is an artificial intelligence (AI) company.
Additionally, it makes sense that every company must now think of itself as a generative AI company.
But how many generative AI competing companies also produce platform-level tools for software application development professionals to create AI-driven, AI-enriched, AI-augmented, or AI-accelerated (other adjectives are available) code that operates at a high level of quality across a significant user base?
This (certainly arguably) smaller definition of how AI know-how can be applied to any company can be applied to Tel Aviv-based CodiumAI, a generative AI company focused on codebase development. The company today announced the deployment of a proactive AI agent for all 500,000 users of its software development tools.
The semi-autonomous solution works in tandem with popular code completion tools to improve the quality of AI-generated code, proactively suggesting improvements to developers as they code. Along with the agent, CodiumAI also publishes its own code completion tool that emphasizes code integrity.
According to CodiumAI, code generation powered by generative artificial intelligence is ‘ubiquitous’ with 92% of developers already using AI tools to help write code.
At the programmers – and…
But as we know, the role of a programmer involves more than just writing code. Creating accurate documentation, improving existing code, and exhaustive testing are cornerstones of the profession, and current AI assistant technology ignores them.
While AI assistants help developers write code faster, there is great concern that they could exacerbate software quality problems.
Codiumate Agent works alongside developers as they work, suggesting relevant tests, creating documentation, and identifying duplicate code without prompting. By offering developer assistance that complements the completion tools, AI Agent enables faster and (theoretically) better code.
Our agency future
Designed for enhance CodiumAI’s existing code integrity and testing offering, the agent brings testing and validation directly into your existing development environment.
“While coding assistants like GitHub Copilot are extremely popular, more and more developers are insisting on capabilities beyond what’s available,” said CodiumAI CEO and co-founder Itamar Friedman. “Developers want AI to help them with the often tedious tasks that are critical to ensuring code quality and correctness, and they go beyond just writing code. We at CodiumAI look to the agent future[ieonewheremoresoftwareagentsareused)wherebusydeveloperswillteamupwithagentstoimprovecodingqualityandprocessesIntegratedagentswillbethereeverystepofthewayofferinginsightsrecommendationsandsolutionsproactively”[tjonuukojojsekoristivišesoftverskihagenata)ukojojćesezaposleniprogrameriudružitisagentimakakobipoboljšalikvalitetukodiranjaiproceseIntegriraniagentibitćetunasvakomkorakuproaktivnonudećiuvidepreporukeirješenja”[ieonewheremoresoftwareagentsareused)wherebusydeveloperswillteamupwithagentstoimprovecodingqualityandprocessesIntegratedagentswillbethereeverystepofthewayofferinginsightsrecommendationsandsolutionsproactively”
This announcement builds on the recent success of CodiumAI’s AlphaCodium research, which it proposed a new approach to code generation by LLM, which is built on a test-based, multi-phase iterative flow for solving code problems. AlphaCodium’s performance on the CodeContests benchmark showed that its performance improved GPT-4’s accuracy from 19% to 44%.
Developers can try the coding agent by installing Codiumate for VS Code or JetBrains IDE, start with a Google search for CodiumAI.
CodiumAI tweets at @CodiumAI
