Overview
Shutterstock (NYSE: SSTK), a global leader in creative content operating at scale across more than 150 countries, sought to enhance the efficiency and accuracy of its incident response process. Shutterstock powers creative workflows for enterprises, media companies, and creators worldwide, providing licensed images, video, music, and AI-generated content, supporting over 1.5 million customers across marketing, media, and digital production. With over 300 engineers and technical operators overseeing a complex, always-on digital platform serving millions of customers, reliability remains critical to customer trust and business performance.
As Shutterstock’s digital infrastructure expanded, incidents increasingly required fast, accurate context and coordination across distributed systems to prevent service degradation and operational disruption. The team turned to Vibe AI to streamline how incidents are identified, analyzed, and communicated across the organization. With Vibe AI, Shutterstock introduced a real-time AI Incident Commander that operates around the clock 24/7, automating detection, coordination, and contextualization while capturing incident knowledge to drive faster resolution and continuous reliability improvements over time.
Challenge
Before implementing Vibe AI, Shutterstock often relied on manual updates and fragmented information sources across traditional incident management tools such as PagerDuty and Jira. While these tools captured alerts and tickets, a significant portion of outage time was spent outside of technical diagnosis on coordination, communication, and context-gathering across people and systems. This made it difficult to maintain a shared, real-time understanding of incidents, particularly across globally distributed teams operating in multiple regions and time zones.
When incidents escalate at Shutterstock, multiple layers of responders are involved: Tier 1 on-call engineers, Tier 2 service owners, incident commanders, and leadership stakeholders, each requiring timely and accurate context. Previously, engineers would be frequently interrupted during active incidents to answer recurring questions like "What's going on?", "What's impacted?", and "Who's working on this?" across Slack channels, bridge calls, tickets, and dashboards. This created delays in identifying root causes and communicating timely updates to executives and the broader user community. Responders often had to juggle between investigating the issue and synthesizing updates for others, increasing cognitive load and slowing overall resolution. As incidents crossed regions, team handoffs frequently introduced communication gaps, inconsistent status updates, and delays in executive visibility.
Over time, this pattern contributed to alert fatigue and responder burnout, as engineers remained on-call not just to solve technical problems but also to act as human routers of information across tools, teams, and stakeholders. The lack of a single, continuously updated source of truth made it difficult for incident commanders and leaders to trust the latest state of an incident without repeated manual check-ins. The Shutterstock team needed a solution that could reduce Mean Time to Detect (MTTD) and Mean Time to Resolve (MTTR) not only by accelerating technical triage but also by removing the coordination, reporting, and visibility gaps that prolonged outages.
Key Outcomes
Reduction in
MTTD/MTTR
Anomaly signals detected in Q4 2025
Decrease in minutesper incident
Accuracy in incident summarization
Solution: Why Vibe AI
By integrating Vibe AI into its incident operations workflow, Shutterstock automated key aspects of incident management during active incidents across detection, triage, coordination, and reporting. Vibe AI acted as a real-time AI Incident Commander, continuously analyzing event data, correlating signals across systems, and surfacing actionable insights to support faster Tier 1 triage and escalation. It continuously gathered and curated current, accurate incident context, supporting more precise escalation routing and enabling Tier 2 and senior responders to focus on deeper analysis rather than manual context gathering.
Instead of ingesting data indiscriminately, Vibe AI operated persistently across the full incident lifecycle, detecting PagerDuty incidents, monitoring incident-related signals and conversations in Slack, and routing escalations to the right teams. It pulled service ownership and dependency context from Backstage to identify impacted services and teams upstream and downstream, while allowing responders to manage related action items directly through Jira. Vibe AI referenced runbooks and operational guidance from Confluence to recommend relevant next steps and correlated alerts, logs, conversations, and historical incident data into a single, continuously updated view.
As incidents unfolded, Vibe AI automatically generated executive-ready incident reports aligned with leadership expectations. These reports captured root cause, response timelines, ownership, and next steps, reducing the need for repeated manual updates while giving leaders a trusted, real-time view of incident status. This improved response efficiency, enabled more consistent incident handling, and contributed to long-term reliability improvements.
Vibe AI transformed incident response from manual coordination to real-time, Al-driven incident command.
The Premium Partnership Experience
The Vibranium team worked closely with Shutterstock to align Vibe AI with how incident response actually runs across a large, globally distributed organization. Rather than introducing new tools or processes, Vibe AI was embedded directly into Shutterstock's existing systems and communication channels. The collaboration emphasized a high-touch, adaptive partnership focused on real-world workflows, on-call operations, and transparent feedback cycles.
By observing live incidents, Vibranium and Shutterstock identified recurring friction beyond technical diagnosis. A common issue involved engineers and incident commanders being asked to create or update tickets mid-response even though the necessary context already existed across alerts, Slack conversations, and bridge calls. This manual overhead disrupted investigation and increased cognitive load during critical moments. In response, Vibe AI was designed to automatically generate and update incident tickets within Shutterstock's workflow, enriching them with real-time context such as signals, ownership, timelines, and key decisions.
The engagement reflected a genuine partnership that converted operational pain points into scalable improvements. Continuous feedback accelerated targeted enhancements, including identifying the right runbooks, refining escalation routes, surfacing historical patterns, and supporting tabletop exercises grounded in past incidents. The result was an experience that felt premium and personalized. Vibe Al made incident response more intuitive for engineers. more effective for commanders, and more transparent for executives, integrating seamlessly with Shutterstock's existing tools and processes while strengthening reliability across the organization.
Beyond live incident support, Vibe AI has become an invaluable tool for building Post-Incident Reviews. Through its integration with our CMDB, it automatically identifies impacted services and areas, produces high-quality documentation, and proposes meaningful remediation actions with clear ownership. As a result, our PIRs now strike a strong balance between technical depth and executive-level clarity, enabling the wider business to engage, understand, and collaborate more effectively on preventative actions.
