Developers Max Brodeur-Urbas and Rahul Behal say AI has the potential to automate many business-related tasks, but many AI-powered automation tools on the market today are unreliable and expensive. We believe that this is the case. Brodeur-Urbas told TechCrunch that part of the problem is that users expect too much from AI. For example, users assume that they can handle highly specialized and niche workloads where accuracy is important.
“If users want to use AI for enterprise purposes, the technology needs to have no margin for error,” Brodeur-Urbas said. “It is not realistic to completely delegate certain workflows to AI. [an AI] Spin your wheels and run the same Google search over and over again. ”
Still, Brodeur Arbus, a former Microsoft software engineer, and Béhar, formerly a software developer at Amazon Web Services, thought AI could have narrower uses today. So they started thinking about how to squeeze what Brodeur-Urbas calls “real value” out of AI technology.
These ideas became a wrapper for the open source app Auto-GPT, then a proof of concept, and eventually the startup Gumloop. Gumloop uses AI to automate repetitive workflows and streamline basic tasks.
“We started the company in our bedroom in Vancouver as a side project,” Brodeur-Arbus said. “We were trying to solve a very simple problem for a group of non-technical people in a Discord server, but it turned out to be a bigger problem than we ever imagined. Ta.”
Gumloop provides a workflow builder that integrates with third-party apps and tools such as GitHub, Gmail, Outlook, and X. Users can build automation by dragging modular components onto the canvas or choose from pre-built pipelines for tasks like generating and summarizing daily inventory reports. document.
Visualized Gumloop's workflow builder. Image credit: Gumloop
Brodeur-Urbas claims that teams at Instacart and Rippling use Gumloop for a variety of use cases.
“Thousands of users now rely on Gumloop as a core tool for their business,” he said. “We found a market appeal in providing non-technical people with tools that allow them to solve problems themselves without relying on engineers.”
There is no shortage of workflow automation tools. Parabola, Tins, Guided AI, and Nanonet come to mind. And we're on the horizon for “agent” tools like OpenAI that promise to automate more complex tasks end-to-end.
To remain agile, Gumloop plans to keep its team fairly small. The company is hiring, but plans to limit the number of employees to 10, Brodeur-Urbas said.
“By using AI for coding, we can achieve throughput equivalent to a team of 20 people and outperform our competitors,” he claimed. “Our plan is to become a billion-dollar company with 10 people.”
As it prepares to move from Vancouver to San Francisco, Gumloop invests in Nexus Ventures with participation from First Round Capital, Y Combinator, and angel investors including Instacart co-founder Max Mullen and Databricks co-founder.・Completed a $17 million Series A round led by Partners. and Chief Architect Reynold Singh. To date, Gumloop has raised $20 million in funding.
“We didn't need any money at all,” Brodeur Urbas said. “It's not about raising money; it's about building a product that people love. This new venture capital will help us build and scale that product even faster.”