From Startup to Acquisition: How We Helped WootCloud Cut Costs by 44% Through Data Engineering and Cloud Optimization

However, their product wasn’t robust enough to fulfill their mission and flag real-time threats effectively. That’s when NaNLABS stepped in. Here’s how we strengthened their cloud-based infrastructure, improved scalability and robustness, and leveraged AI/ML models to enhance data processing and build real-time data pipelines.

Table of contents

• The client: An IoT device management company that needed to scale

• Key challenges: A buggy, costly, and unreliable platform that couldn’t process data in real-time

• Solutions: Improving data processing and infrastructure robustness

• Results: A reliable, scalable, and robust application 44% less expensive than expected 

• What followed: A strategic acquisition by an industry leader

The client: An IoT device management company that needed to scale

WootCloud is a pioneer in enterprise Internet of Things (IoT) device security. It uses device context on its risk assessment platform to scan radio and networks, uncover all unmanaged devices, and apply Zero Trust principles. This ensures all smart devices are properly identified, authorized, and validated when they enter your network.

As you can imagine, each IoT device generates tons of data, such as device activity, network traffic, and security logs—streaming at high speed. WootCloud needed to be robust and scalable to stream, process, and analyze this information in real time, which wasn’t when we started working together.

Enter, NaNALBS: A cloud and data-engineering expert

Our past collaboration with some of WootCloud's lead engineers on a different project helped establish a foundation of trust. They were confident in our ability to integrate seamlessly into their team and work as partners, which was exactly what they needed.

On top of that, WootCloud brought us on board for our deep expertise in cloud data engineering, particularly within the cybersecurity industry.

Key challenges: A buggy, costly, and unreliable platform that couldn’t process data in real time

WootCloud aimed to reach enterprise-level companies, but when we started working together, it wasn’t ready to take on a big influx of data.

The main challenges this client faced include: 

• Slow data processing and response: WootCloud had an inefficient process for aggregating raw data in a central location. This complicated real-time data analysis and response to security events, such as immediate threat detection and remediation.

• An unscalable product due to buggy code: WootCloud relied on EC2 instances which were difficult to maintain due to frequent patching, upgrading, and manual scaling, leading to a poor development experience. The code was also highly coupled and didn’t go through testing or QA controls.

• Inefficient internal processes: The existing development process was limiting team velocity and delaying time to market. Engineers were working in silos and had no processes in place or team guidelines to follow. This caused them to constantly be putting out avoidable fires because they didn’t have time to make improvements or fix the root cause of problems.

• Functional UI/UX: WootCloud wanted to develop new features, make them a seamless part of the platform, and improve the UI/UX flow.

• Expensive infrastructure and servers: Since the development process wasn’t standardized and the platform relied on EC2 instances, WootCloud was spending unnecessary extra money.

Facing similar challenges? As your tech sidekick, we'll help you tackle them head-on. Let's talk!

Get in touch

Solutions: Improving data processing and infrastructure robustness

One of the core function of WootCloud's platform is to analyze device data and trigger real-time threat alerts. Since the platform wasn’t scalable or reliable as it was, it couldn’t gather, process, and analyze the data fast enough.

This led us to:

• Adopt a serverless architecture using AWS Lambda. As mentioned, WootCloud relied on EC2 instances which led to a poor development experience and high costs. By migrating to AWS Lambda, we removed the need for manual server management and enabled automatic scaling based on workload demand. Take a look at the diagram illustrating the new architecture:

• Use AWS services for better resource management. Many AWS services support auto-scaling, which reduces operational overhead and improves cost efficiency. This way, we helped WootCloud build a highly responsive, cloud-native solution capable of handling massive data loads without compromising performance. In this case, we used:

  • AWS Lambda to allow the platform to scale automatically and execute code only when needed.

  • AWS S3, for secure and scalable storage and to make data always accessible.

  • AWS Elasticsearch to enhance real-time search and analytics, enabling fast querying and clear data visualization of threats.

• Build a scalable database infrastructure with MongoDB and Elasticsearch. This simplified data retrieval, storage, and processing for time-sensitive information. Using MongoDB as a flexible and scalable NoSQL database to gather and store device-related information, allowed for dynamic schema design without causing bottlenecks. We used Elasticsearch to index and search security events, enabling real-time access to critical data for alarm triggering.

• Set data pipelines for real-time large data processing. We used Apache Spark to aggregate and analyze big datasets. It processes raw data from IoT devices in real time while extracting meaningful insights and identifying patterns or anomalies. This was integrated with Apache Kafka to ensure a continuous data flow between storage, search, and analysis components. 

• Integrate AI/ML models for threat prevention. We applied machine learning models for device classification and anomaly detection. We also integrated machine learning models to recognize unusual behavior and flag potential security threats before they escalate.

• Set automated testing to improve the platform’s reliability. We used pytest to ensure high test coverage for Python APIs, reducing system failures and regression issues. We also used Jest for unit testing, SonarQube for continuous code quality checks, and Cypress for end-to-end (E2E) testing.

• Redesign the UI/UX. We used the atomic design methodology to reuse components from these five stages: atoms, molecules, organisms, templates, and pages. We also included a deep linking method to simplify the user experience (UX) and make it easier for users to navigate the app.

Ready to leverage the power of cloud-native solutions? Let's talk!

Get in touch

Tech stack

To help the client achieve their goals, the NaNLABS squad harnessed the following tools and technologies:

Results: A reliable, scalable, and robust application 44% less expensive than expected

Our partnership with WootCloud drove impactful improvements across the board, including a significant reduction in operational overhead. Key outcomes include:

• 40% reduction in data processing times: Optimizing data processing workflows allowed WootCloud to analyze security events faster than ever before. This speed boost means threats are now detected and mitigated almost instantly, keeping WootCloud customers ahead of potential cyber risks. 

• 25% increase in anomaly detection accuracy: By fine-tuning machine learning models, we reduced false positives significantly and guaranteed the system only flags real threats. We also streamlined data workflows, which reduced processing latency and allowed the system to quickly detect and respond to potential threats.

• 44% reduction in server bills: Shifting WootCloud to a serverless architecture and minimizing server idle time led to a significant reduction in monthly costs. WootCloud CTO shares: “We weren’t expecting more than a 20% reduction, but when we saw the final numbers, the bill was 44% lower. It was a game-changer.” 

• Improved team velocity: Eliminating manual server maintenance, patching, and scaling significantly simplified the development experience. This freed up the team to work more efficiently. For instance, DevOps teams can now focus on innovation instead of handling and upgrading EC2 instances. This also made WootCloud’s platform more scalable, cost-effective, and developer-friendly.

• 3x reduction in debugging time: With automated testing and additional support, the team quickly identified root causes, minimizing the time engineers spent resolving recurring issues.

What followed: A strategic acquisition by an industry leader

When we first started working with WootCloud, the platform was in the early stages. It had a market-revolutionizing idea but needed to be more robust to serve enterprise-level clients.

After four years of collaboration, WootCloud was acquired by Netskope, the global leader in Security Service Edge (SSE) and Zero Trust solutions, in 2022.

This success story demonstrates how strategic technical improvements and collaborative efforts can lead to significant business outcomes, such as revenue growth and successful acquisitions.

Ready to scale your business with cloud-native solutions that drive real results? Let's talk!

Get in touch

Editorial contributions: Camila Mirabal