How ThetaRay Is Rebuilding Its 24/7 NOC with Logz.io and OrionIQ

ThetaRay is a global fintech AI company using proprietary machine-learning models to detect financial anomalies for banks and fintechs worldwide. Today the company operates both as an on Premise and Saas hosted via Azure cloud supporting real time payment infrastructure where downtime tolerance is measured in seconds.

As ThetaRay scaled from software delivery into a SaaS operation, observability quickly became mission critical.

The Observability Challenge

ThetaRay’s infrastructure runs on a large Kubernetes-based microservices environment spread across dozens of customer deployments.

Before Logz.io:

• Engineers manually searched through pods during incidents
• Visibility across environments was fragmented
• Troubleshooting became slower as the platform scaled
• RCA investigations required heavy manual correlation work

When services restarted unexpectedly, engineers were forced to jump from pod to pod searching for the failing instance, manually correlating logs across environments as the fleet continued to grow.

For ThetaRay, the stakes were even higher than typical SaaS operations. Every outage is also a regulated event. The company must produce formal RCA reports for customers and regulators after incidents, making fast and accurate investigations critical for both engineering and compliance.

“Before Logz.io, troubleshooting meant jumping between pods and environments manually. Today we investigate incidents from a single place and resolve them dramatically faster.”

Yossi Cohen
Senior AI & Operations Engineer, ThetaRay

The Solution

ThetaRay consolidated observability on Logz.io, centralizing metrics, logs, and audit-layer application events.

The platform provided unified visibility across all customer environments, allowing teams to monitor fleet-wide health from a single location instead of investigating issues one environment at a time.

The impact was immediate, transforming the day-to-day work for NOC operators. With unified visibility across all customer environments, operators could monitor fleet-wide health from a single location, eliminating the need to manually jump between pods and correlate logs during an incident. This centralized investigation dramatically accelerated troubleshooting time from hours to minutes, improved incident response consistency, and reduced manual operational overhead.

That observability foundation enabled ThetaRay to establish a dedicated 24/7 NOC operation supporting around-the-clock monitoring and response.

The NOC is often staffed by a single operator managing a constant stream of alerts across multiple screens and systems. With Logz.io providing centralized visibility and operational clarity, ThetaRay significantly improved its ability to detect, investigate, and respond to incidents in real time.

ThetaRay operators went from jumping across pods and environments to investigating from a single pane, almost immediately.

Accelerating RCA with Logz.io 360 AI

On top of centralized observability, ThetaRay adopted Logz.io’s AI-powered Root Cause Analysis (RCA) capabilities.

Previously, RCA generation required engineers to manually correlate logs and services during high-pressure production incidents. Today, investigations are significantly faster and more consistent through guided AI-assisted workflows.

For a regulated fintech organization, this directly improves the company’s ability to deliver timely RCA documentation to customers and regulators. 

“For us, RCA is not just engineering. It’s a regulatory requirement. OrionIQ helps us move faster and more consistently under pressure.” – Yossi Cohen

OrionIQ for NOC Automation

ThetaRay is now piloting OrionIQ as a design partner.

The AI agent operates on the same playbooks used by human NOC operators today, reasoning over telemetry already managed inside Logz.io to perform end-to-end investigations. In one recent investigation, after a runbook change instructed OrionIQ to investigate deeper, the agent identified the correct root cause after the on-shift engineer initially dismissed it as unrelated.

During recent P1 incidents, OrionIQ:

• Accelerated RCA generation
• Surfaced findings missed by human operator
• Reduced troubleshooting time significantly

“OrionIQ gave us the foundation to establish something we’d needed for a long time: the 24/7 NOC Automation.” – Yossi Cohen

Why ThetaRay Chose Logz.io and OrionIQ

ThetaRay cites two primary reasons for choosing Logz.io.

Technical Fit

The platform provided the depth, dashboarding, and observability consolidation needed to support a financial-grade SaaS operation at scale.

Strategic Partnership

From its early stages, ThetaRay valued working closely with its technical vendor. Over time, the relationship evolved into what Yossi describes as a co-development partnership rather than a traditional vendor relationship.

“Logz.io is intuitive. It learns to speak your language. The user experience became second nature for our team.” – Yossi Cohen

The Future of NOC Operations

ThetaRay’s long-term vision is to evolve from a fully staffed 24/7 NOC into a largely autonomous operation powered by OrionIQ.

The company’s goals include:

• Automating repetitive low-priority incidents
• Increasing internal resolution rates from 70% to 90%
• Allowing engineers to focus on higher-value SRE work
• Reducing operational overhead without reducing team size

P4 incidents currently represent roughly 45% of ThetaRay’s ~1,000 monthly support cases, making automation a significant operational opportunity.

ThetaRay’s goal is not to reduce headcount, but to elevate the role of the operations team from repetitive shift-based work into higher-value engineering ownership.

“We stay at 20 people, but those people move up. From low-skill, shift-based work to higher-skill SRE work managing the parts OrionIQ doesn’t.” – Yossi Cohen