Salesforce CodeGen Unpacked: Written English Phrases Become Executable Code with Conversational AI
The Salesforce Research team has lifted the veil on CodeGen – a new, large-scale language model built on the concept of conversational AI programming.
What’s the impact: Using Salesforce’s CodeGen model, business users and developers can use natural language to describe what they want the code to do — like build an app that surfaces the latest customer interaction. The machine then translates those English prompts into executable coding, essentially, writing the app for them using only written commands.
Driving the news: CodeGen’s conversational AI capabilities help anyone avoid the time-consuming process of building apps from scratch. Much like low code, CodeGen empowers users to develop apps quickly without extensive coding, opening up more time for complex tasks that benefit from a human touch.
The Salesforce perspective: “The future of coding is the intersection of human and computer languages — and conversational AI is the perfect bridge to connect the two,” said Silvio Savarese, EVP & Chief Scientist, AI Research, Salesforce. “With models like CodeGen, conversational programming is becoming more of a reality and mainstream.”
Zoom out: As the world becomes increasingly digital-first, the tech industry needs to answer an important question: “What happens when future programming needs become so complex that the skills needed outstrip human capabilities?”
Conversational AI models like CodeGen are uniquely positioned to solve this problem by combining human insights and creativity with AI-driven coding capabilities. That intersection is poised to create innovative new solutions to the biggest technology challenges at scale.
What’s next: Dive deeper into CodeGen here, and find more Salesforce Research insights on the team’s AI and Research blog.
