How an AI company used Gatling Enterprise to enhance performance infrastructure

Why Gatling Enterprise?

  • ExploreAI aimed to handle 10,000 to 12,000 users concurrently interacting with their system
  • Reduce their response time which is currently 3000 to 5000 milliseconds
  • Ensure scalability of their product
    Commence load-testing with Gatling Enterprise
  • Reduced response time to 150-300 milliseconds
  • Handle 180,000+ users monthly, which is 10x the original capacity
  • Next goal to run Gatling Enterprise in CI/CD pipeline

About ExploreAI

ExploreAI founded in 2013 with the mission is to enhance the progress of its clients' digital teams by providing services such as workforce training, talent recruitment, and sponsorship for students enrolled in a data science program. These solutions are driven by the ExploreAI Academy, an educational institution with award-winning scientists and engineers dedicated to imparting data and AI skills to empower the future generation and scale AI into the organization to change how the business is run.


Location: South Africa

Industry: Software

Turnover: $30.6 million + (2022)

Employees: 185 (2023)

Business Metrics: Average 180,000 website visits per month

Gatling Enterprise Users: Software Engineer and results shared within the team

Start Your Success Story Start your free trial, see what Gatling can do for your team, and enhance your performance engineering.

"Gatling is the first external provider that we’ve worked with that’s been this focused on interacting with their clients which is very refreshing and helpful. Gatling takes the load out of Load Testing."
Justin Bedser

Software Engineer, ExploreAI


ExploreAI initially conducted load-testing themselves by writing scripts when they heard about Gatling Enterprise from a reference. ExploreAI was introduced to load-testing by Gatling Enterprise.

With Gatling Enterprise, ExploreAI aimed to focus on their performance enhancement efforts on optimizing their Apache stack. This involved fine-tuning configuration settings for request handling, adjusting server parameters, optimizing memory allocation, and vertically scaling the machine to accommodate heavier workloads. 

ExploreAI works extensively on their database and aims to optimize indexes through approximately  40 to 50 distinct queries. Also, the company aims to decrease their current response time, which ranges from 3,000 to 5,000 milliseconds.


Despite having no prior proper experience in load testing with a tool, the ExploreAI team successfully integrated Gatling Enterprise into their workflow with minimal learning curve in February 2023.

ExploreAI achieved this by leveraging Gatling Academy's resources, documentation, and external references like Gatling Community, enabling them to harness the full potential of Gatling Enterprise. With the Gatling Enterprise support team, the integration process was easy, and the company obtained the proof of concept (POC) quickly, facilitating a compelling demonstration to ExploreAI's decision-makers.


With continuous load-testing and Gatling Enterprise's integration, ExploreAI reduced their response times to 200-300 milliseconds in a remarkably short timeframe. Since incorporating Gatling Enterprise into their testing process, ExploreAI now boasts the capability to handle ten times the volume of user traffic compared to their initial capacity. ExploreAI can smoothly support around 180,000 users monthly. 

The substantial success experienced by ExploreAI has paved the way for future expansion. ExploreAI plans to extend the usage of Gatling Enterprise and incorporate load testing into their CI/CD pipeline, further enhancing their software development and deployment processes.

Related Articles

From Our Blog

Stay up to date with what is new in our industry, learn more about the upcoming products and events.

Test faster with enhanced no-code test-building capabilities

Test faster with enhanced no-code test-building capabilities

Jul 3, 2024 10:59:58 AM 4 min read
Putting SSE to Work: A Load Test LLM API Case Study
LLM blog banner

Putting SSE to Work: A Load Test LLM API Case Study

Jun 11, 2024 10:31:41 AM 3 min read
Gatling doubles its availability with a new JavaScript SDK
Javascript SDK

Gatling doubles its availability with a new JavaScript SDK

May 23, 2024 2:56:30 PM 4 min read