AI startup Swa Technology today released a multi-model AI orchestration platform that integrates models from open source and commercial providers to meet developers where they are and reduce shadow AI.
While many organizations try to demystify IT and AI by offering users a limited choice of tools, the Swa platform provides secure, permissioned access to all popular and widely used AI models currently available.
“If workers aren’t given the tools they want, they’re going to go get them,” Mike Sirchuk, founder and CEO of Swa, told the SD Times. “Basically, we’re trying to solve the question of which AI do I choose? And you don’t have to choose one. You just get them all.”
What Swa does is, it matches the user’s input with the model that best fits the query. “Some models do better than others, depending on what the question is,” he explained. “Some models you can’t ask math questions because it will make them very wrong. Some models can do math questions very well. Some models are more conversational and good for marketing, and some are more logical and good for code work flow and designs.” Also, the platform may detect that the model is down, so it will route the request to a different model.
The platform separates the context layer from customer information and brings AI agents to productivity tools like Slack, Zoom and text messaging, for example, allowing companies to build a security layer around it so they can manage who has access to which model, what data each model has and how much data is stored. Sirchuk noted that nothing is ever trained in Swa’s systems, giving the platform the ability to switch and select from any AI provider at any time.
Businesses can upload to the Swa administration portal their proprietary information, such as marketing information, product information, company information, PTO policies, expense report policies and the like. “So a new employee or an existing employee, instead of talking to someone, they just ask, ‘Hey, what’s our PTO policy?’ They will get the answer directly from the policy. ‘Swa, I have an expense report. Here is the receipt. Can you file this expense report for me?’ Yes, it can export the receipt information to this expense report policy and send it to the required policy stakeholders.” This, says Sirchuk, removes many layers of IT management from many different software applications.
Swa, a Sanskrit word meaning ‘self,’ describes the company’s all-in-one approach to AI. The company was founded 18 months ago by engineers who were laid off from Nike, who took their knowledge of building bots. Sirchuk said Nike used bots to detect when the system went down and then restart it. “Well, you throw a productive AI into those bots, and they can do more than just reset the scene. They can see things that they weren’t programmed to see before. They can do actions that they weren’t specifically told to do. So after we got together, and we were thinking about what we could do with this technology. The first problem I had was very different, the two or three I had were a lot of cost. I mean, I’m spending 60 bucks per person on just three models.
“And that’s when we’re like, Yeah, this is something. Wow.”
