Tektonic AI Secures $10M Funding to Develop GenAI Agents for Streamlining Business Operations

A few years ago, 'robotic process automation' (RPA) was a buzzworthy topic in enterprise software. Unfortunately, many of these services aimed at automating repetitive business tasks did not fully deliver on their promise. However, the emergence of generative AI may be the missing element needed to enhance these systems.

Seattle-based startup Tektonic is making waves in this field. After emerging from stealth mode, the company has announced its $10 million seed funding round, which is led by Point72 Ventures and Madrona Ventures.

Tektonic’s innovative approach allows users to interact with generative AI agents through natural language to create workflow automation. One key area the company focuses on is the management of quotes and renewals—processes that often demand numerous manual tasks that are difficult to automate due to their unique and evolving nature across different businesses.

Co-founded by Nic Surpatanu, who previously held key roles at Tanium, UiPath, and Microsoft, Tektonic is poised to revolutionize automation. Surpatanu explains, “Last year, generative AI emerged, and it became clear that it could unlock software possibilities that were previously out of reach. Drawing from my experience at UiPath and Microsoft, I understand the limits of traditional automation.”

Surpatanu emphasizes that generative AI should not be viewed as a "magic box." “To truly maximize its potential, it must be integrated with traditional symbolic methods and software,” he notes. This integration aims to enhance adaptability and context awareness within automation systems, addressing challenges that RPA solutions often face. Traditional RPA tools typically falter when significant changes to the user interface occur, necessitating ongoing maintenance of scripted automations.

Generative AI also facilitates the extraction of semantic entities, helping to align user intent with automation processes. “Our approach won’t guarantee 100% flexibility, but it introduces sufficient adaptability to manage a wider variety of scenarios than ever before,” Surpatanu remarks. While he believes current models aren't yet reliable enough for fully autonomous systems, he anticipates that tools like Tektonic can improve the automation of processes from 50% to 80%, representing a significant advancement.

On the technical front, Tektonic integrates foundation models and open models for entity extraction and basic actions. “Instead of performing manual tasks across various applications, sales reps should collaborate with AI agents who understand their workflows, allowing them to focus on customer engagement,” noted Steve Singh, Ted Kummert, and Palak Goel of Madrona in their announcement. “The rise of generative AI models capable of reasoning across data silos in applications will enable us to rethink process automation, pushing it to new frontiers.”

Currently, Tektonic is in its early stages and is working with design partners to test and refine its system. “Looking ahead three to five years, we envision becoming a SaaS company, seamlessly connecting with the APIs in your systems,” Surpatanu states. In the meantime, businesses can begin using Tektonic by installing the system as a container in their virtual private cloud.