Founded by tech leaders with multiple successes and extensive expertise in AI and cybersecurity at companies such as Google, F5, and Palo Alto Networks, Exaforce has developed the industry’s first multi-model AI for solving challenges in security and operations. This new approach blends large language models (LLMs) with semantic and behavioral models into a powerful AI engine that unlocks unprecedented accuracy, repeatability, and productivity for the SOC.
“We believe Exaforce’s multi-model approach is unique in the industry and will dramatically reduce the false positives and investigation times we experience in our cloud and SaaS environments,” said Pranay Anand, Vice President at NTT Data. “The platform augments our SOC teams by delivering streamlined security operations and faster incident response for every client, freeing up more time to focus on proactive threat hunting.”
Enterprises are asking for a SOC solution that is better at delivering effective and consistent response to threats, faster at detecting and investigating issues, and cheaper to scale defenses on demand without scaling people.
SOC analysts face a deluge of alerts—most of which are false positives—leaving them burdened with massive datasets and manual tasks like log stitching, user validation, and ticket management, which drain resources and slow response times. Detection engineers, meanwhile, struggle with the threat coverage for cloud environments where native threat detection is often lacking and traditional SIEMs offer inadequate coverage. This forces them to write and maintain complex sets of SQL/python code, yet major detection gaps remain. Simultaneously, threat hunters are mired in manual, time-consuming workflows that impede proactive threat detection, making it difficult to stay ahead of attackers.
The well-documented shortage of skilled security professionals exacerbates these issues, making it difficult for organizations to maintain expertise across all SOC roles. As a result, SOCs risk burnout, delayed response, and increased exposure to growing threats.
Solving these challenges requires an innovative team with concentrated backgrounds in cybersecurity, AI, and cloud operations. Exaforce’s founding team unites expertise across all three – with firsthand experience leading the world’s most complex SOC environments.
They’ve operated large scale security services at F5, protecting the world’s biggest banks and social networks, designed the complex models underlying Google’s AI services, spearheaded the industry-leading cloud security platform at Palo Alto Networks, and successfully founded and scaled multiple startups. Through these experiences, Exaforce’s founding team has gained a front-row seat to the problems faced in today’s SOC and a nuanced grasp of AI’s potential and pitfalls.
“At Mayfield, we invest in founders first and foremost, which is why we backed Exaforce at the ideation stage in our third collaboration with Ankur Singla,” said Navin Chaddha, Managing Partner at Mayfield. “What excites us is how Exaforce is reimagining the massive opportunity of developing AI teammates to offload complex tasks that help humans increase productivity and efficacy, and they are starting with the SOC market where the problems of skills and talent are acute. The team’s progress since those initial whiteboard sessions—securing a dozen enterprise design partners, and delivering 10x improvements in SOC efficiency—validates our early conviction that Exaforce is building something revolutionary in the collaborative intelligence era.”
The right AI solution for the SOC must analyze enormous volumes of logs, cloud telemetry, and threat data to make rapid, high-stakes decisions. Agentic solutions that rely solely on LLMs can only review a fraction of that data at once, resulting in incomplete problem analysis and reasoning that is unreliable and hallucination-prone.
Exaforce overcomes this technical barrier with a multi-model (aka layered) AI engine that is purpose-built for security and operations. It applies these models in combination, starting with a semantic data model, along with statistical and behavioral models, to extract key insights, behaviors, and relationships from raw data, then performs deeper analysis with knowledge models. This structured use of multiple models not only rightsizes, but also improves the quality of SOC data, which is then fed into LLMs—enabling end-to-end reasoning across the full scope of data. This approach avoids the blind spots of systems that use only LLMs and delivers more accurate, repeatable results.
While in stealth, the company has been collaborating with over 10 leading enterprises across technology, AI software, energy, and manufacturing markets to refine this multi-model approach, and delivering significant gains to their SOC teams in the process.
“Our vision is to empower SOC teams with an intelligent platform that allows humans to collaborate seamlessly with AI agents—integrating precise human oversight with advanced automation,” said Ankur Singla, co-founder and CEO of Exaforce. “By harnessing Exabots alongside advanced data exploration, we reduce false positives and eliminate tedious busywork—all while giving humans the flexibility to choose where they want to be hands-on. This approach has already delivered 10x improvements to our design partners. Our design partners have already seen tenfold improvements using this approach.”
