Securing the next generation of cloud and AI workloads with Upwind

As cloud and AI adoption have rapidly increased over the past decade, infrastructure has transitioned into a new chapter — from static servers to a dynamic fabric of containers, microservices, and managed services. Early cloud security tools were built for that first chapter, focusing on posture management, or taking static snapshots of assets and scanning configurations and permissions to help teams simply understand their attack surface. These tools treated dormant vulnerabilities in unused services with the same urgency as issues on hot production endpoints, flooding security and DevOps teams with endless alerts. With little ability to distinguish signal from noise, teams have struggled to understand which vulnerabilities are truly pressing. Today, CISOs and security leaders are converging on a new realization: in modern cloud environments, runtime context is key. 

Upwind is building the Cloud Native Application Protection Platform (CNAPP) for today’s containerized, API-driven cloud infrastructure, where workloads spin up and down in seconds, and dependency graphs constantly shift. Upwind’s inside-out approach starts inside workloads, where its lightweight sensors capture how applications are running. These sensors turn network, process, and container events into a continuous data stream. But Upwind’s real advantage lies in its correlation engine, which enriches these streams with metadata, identities, and the context necessary to reconstruct an actual execution graph of the environment. From there, the platform has a clear line of sight into whether a vulnerability is truly reachable within a specific container or service, and what an end-to-end attack path may be.

Why we’re partnering with Upwind

At Bessemer, we’re excited to back the team at Upwind by leading their $250 million Series B. Beyond making strides in a competitive landscape with established incumbents, Upwind exhibited three of our core beliefs on success in the current landscape:

• Runtime-first in the container era: With containers and Kubernetes now the standard architecture, deep runtime visibility is becoming a requirement for every CISO. Upwind is fundamentally developed with runtime at its core, rather than treating it as a bolt-on feature to a legacy architecture
• Signal over noise: While capturing raw sensor data is relatively easy, transforming it into actionable, context-rich insights is a harder problem to solve. Upwind’s correlation engine and execution graph filter out background noise and elevate vulnerabilities and behaviors that map to truly exploitable paths, reducing ticket volume while driving measurable improvements in an enterprise’s security posture
• A battle-tested team: Founders Amiram Shachar (CEO), Tal Zur (CTO), and Lavi Ferdman (SVP Growth) are approaching a crowded cybersecurity market with the fresh perspective of seasoned DevOps engineers. Along with their multiple successes across a decade of working together, they continue to impress with a deeply customer-centric mentality and the energy to run through brick walls seven days a week. 

Cloud security built for the AI and real-time era

In one of the largest and most competitive segments of cybersecurity, Upwind stood out to us as the only player that’s truly architected for where cloud security is going, not where it’s historically been. With a product loved by both security and DevOps teams — and actively used as a true platform — we’re proud to partner with the team at Upwind as they build the runtime-first cloud security platform for the next generation of enterprises.