Skip to main content
Stack AI
Stack AI: No Machine Learning Engineers? No problem.

GenAI adoption is outpacing any prior technology platform transition. When OpenAI released ChatGPT in Nov 2022, the consumer-facing application reached 1 million users in only 5 days — something that took Facebook 10 months to achieve. That consumer excitement quickly transitioned to businesses, with Ramp suggesting recently that AI spending is growing at 293% among their customers. Engineers, analysts, lawyers, sales & support teams, and healthcare professionals are all finding ways to improve their day-to-day workflows with the help of AI. The early success cases are significant. Klarna recently reported that they are handling two-thirds of their customer service chats with AI.

As organizations allocate budget and move quickly to begin building genAI capabilities, they all face two significant challenges: 1) a shortage of in-house machine learning expertise available to build these capabilities, and 2) a requirement to adopt an entirely new set of developer tools and infrastructure to power the technology. When we met Bernardo Aceituno and Toni Rosinol in 2023, they were packaging their AI expertise from their PHDs at MIT to simplify the AI app development process and lower the barrier to entry.

Stack AI reduces the complexity of building genAI applications and makes them accessible to businesses everywhere, no matter their technical stack or machine learning expertise. Prior to Stack AI, a developer wanting to build a chat interface for their marketing team’s data might have to select a vector database (e.g., Pinecone, Chroma) and host it somewhere, learn a new AI framework (e.g., Langchain), select an LLM and inference provider (e.g., Google, TogetherAI, OpenAI), choose an embeddings model (e.g., through Hugging Face), develop integrations with key data sources the developer wants to enable access to (e.g., Hubspot, Snowflake, Salesforce), and implement the right auth/permissioning around the application (e.g., Okta). If that sounds complex and arduous to you, many developers would agree. Stack AI simplifies this process into a convenient, GUI-based workflow in which users can select technical components and chain them together to build a wide range of apps for their organization. To get a head start, users can leverage one of Stack AI’s many templates at the click of the button and immediately see the results of their genAI workflow.

Stack AI’s platform becomes even more powerful for users as foundational models and the surrounding technology ecosystem improves. The team has already started to incorporate early multi-modal capabilities and is regularly adding new integrations (e.g., Zapier, Make.com, Airtable) and data sources or plug-ins (e.g., MongoDB, Alogolia, WolframAlpha, Shopify).

We’ve been blown away by the ingenuity and creativity of Stack AI users. They’ve built applications that intelligently craft RFP responses, draft emails to patients at a hospital, and automate a wide array of reporting across functions in sales, marketing, ops, and more. One organization built an AI-powered employee staffing app that stores employees' resumes as vector embeddings, and then selects the best employees for a given job given specific skill requirements. The platform is easy to learn, so even non-technical users can build on Stack AI.

We’re thrilled to have partnered with Bernardo and Toni at Stack AI early in the genAI revolution, and continue to be excited about the possibilities for businesses everywhere. Today, over 40,000 users are building automations and applications on the Stack AI platform. Their users span technical and non-technical thinkers and an ever-increasing set of use cases. We look forward to seeing what everybody builds next.