Goldman Sachs Just Hired an AI Software Engineer

Wall Street's relationship with artificial intelligence just took a dramatic turn. Goldman Sachs made headlines by bringing on Devin, an AI software engineer created by start-up Cognition, marking what could be the finance sector's boldest move into autonomous AI territory.

The investment giant isn't just experimenting. They're going all-in.

Goldman CIO Marco Argenti told CNBC that the company plans to "augment" its workforce with the AI tool, which will execute tasks on behalf of its more than 10,000 human software developers. "Initially, we will have hundreds of Devins, and that might go into the thousands, depending on the use cases," Argenti said. This represents more than gradual adoption. While Wall Street has been testing AI tools for years, Goldman's embrace of autonomous AI agents signals a fundamental shift in how financial institutions view artificial intelligence.

Meet Devin: The World's First AI Software Engineer

Cognition introduced Devin last year with bold claims. The company marketed it as the world's "first AI software engineer," designed to execute software engineering tasks independently while making thousands of decisions along the way. The concept sounds straightforward. Using natural language prompts, programmers can assign Devin complex tasks like building applications or hunting down bugs in existing code. The AI works through problems systematically, much like a human engineer would.

Cognition describes Devin as "a tireless, skilled teammate, equally ready to build alongside you or independently complete tasks for you to review." The description strikes at the heart of modern workplace anxiety; will AI replace human workers or simply make them more productive?

The start-up’s timing couldn't be better. Shortly after Devin's launch, the founder of prominent AI search engine Perplexity praised Cognition's creation as the first AI agent "that seems to cross the threshold of what is human level and works reliably."

Investors took notice. Cognition reached a $4 billion valuation in March after raising hundreds of millions in investment led by 8VC, an early-stage venture capital firm founded by Palantir cofounder Joe Lonsdale.

Marketing rarely matches reality in the AI space. When independent data scientists tested Devin's capabilities, the results told a different story. The AI software engineer successfully completed only three of 20 proposed tasks. Fourteen of the test projects failed completely. While Devin performed twice as well as standard large language model-based chatbots in benchmark tests, it fell short of the promise of full autonomy.

AI-generated code brings its own set of challenges. Bugs can lead to costly downtime. Security risks multiply when human oversight decreases. The technology shows promise but requires careful implementation.

Understanding Agentic AI

Agentic AI represents a category of artificial intelligence tools. Unlike chatbots or research assistants that respond to user queries, agentic AI executes tasks and makes decisions independently. The key difference lies in autonomy. While generative AI focuses on creating outputs, writing emails, compiling research, generating reports, agentic AI takes action. These systems pursue set goals and complete various tasks to achieve them, working more like humans do. The technology's ability to handle multistep jobs from start to finish sets it apart from earlier AI tools. The question remains, how effectively can these systems operate without constant human guidance?

Goldman Sachs appears to be the first major bank implementing a comprehensive AI-coding tool like Devin. Their competitors likely aren't far behind. AI assistants and chatbots already permeate Wall Street operations. Major banks seek competitive advantages through AI-related hires and technology investments. JPMorgan Chase introduced its own generative AI assistant internally last year, making the tool available to 60,000 employees. Workers can use it to write emails and file reports. Morgan Stanley similarly provided financial advisers with internal tools powered by OpenAI's technology.

JPMorgan Chase CEO Jamie Dimon expressed bullish sentiment about AI's potential last year: "We are completely convinced the consequences will be extraordinary and possibly as transformational as some of the major technological inventions of the past several hundred years." He expects the technology to "augment virtually every job." While AI chatbots approach widespread adoption, agentic AI remains in early stages. Companies proceed cautiously due to the technology's relative lack of human oversight. Deloitte predicted that a quarter of companies already using generative AI would begin exploring agentic AI in 2025. The risks associated with reduced human supervision mean organisations will implement these tools gradually.

What This Means for the Future

Goldman's move raises critical questions about the future of work in finance. Will AI tools like Devin replace human software engineers or simply make them more productive? The answer likely depends on implementation. Goldman's approach suggests they view AI as a force multiplier rather than a replacement. The bank plans to deploy hundreds, potentially thousands of AI agents alongside their human workforce.

This strategy could reshape how financial institutions approach software development. Tasks that once required human engineers might become automated, freeing up skilled workers for more complex projects. The implications extend beyond Goldman Sachs. As one of Wall Street's most influential institutions, their embrace of agentic AI could accelerate adoption across the financial sector.

Other banks will watch closely. If Goldman's experiment succeeds, expect rapid expansion of similar programs throughout the industry. If it fails, the financial sector may take a more cautious approach to autonomous AI implementation.

The stakes couldn't be higher. Wall Street's relationship with artificial intelligence is evolving from cautious experimentation to bold implementation. Goldman Sachs just made the first major move in what could become the defining technology battle of the next decade.