How TierZero's AI teammate identified a critical silent bug saving over $150,000 and transformed Eaze's engineering culture.
Eaze is an online marketplace and technology platform that helps provide legal access to cannabis through safe and convenient delivery. Unlike its competitors, Eaze builds and maintains its entire technology stack in-house, giving it a unique advantage in a highly-regulated industry.
With over 5,000 employees, 40 native software applications, and an infrastructure built up over 9 years, keeping everything reliable, compliant, and scalable is a constant challenge. The complexity of such a large in-house ecosystem means even small oversights can cascade into costly, hidden issues.
Before TierZero, Eaze's small engineer team was juggling over 40 native software applications alongside multiple telemetry aggregation and monitoring tools all while firefighting urgent incidents. With limited resources, they often had to make difficult trade-offs between resolving critical issues or training the engineers on the various complex systems at hand.
However, the most dangerous issues may be the silent incidents that do not raise the alarms; in this case, a silent infrastructure flaw quietly costing Eaze over $150,000.
Soon after TierZero was adopted, the powerful AI teammate got to work surfacing abnormalities and responding to alerts. TierZero quickly noticed the undetected issue – something no existing tool or human process had identified. It sounded the alarm, allowing Eaze's engineering team to tackle a problem they didn't know existed – and saving Eaze an immeasurable amount of money.
Now, TierZero automatically analyzes incidents, flags hidden risks, and ultimately frees Eaze's engineers to focus on strategic priorities; it has transformed how Eaze thinks about AI, the health of its infrastructure, and the future of engineering.
Eaze is a company that prides itself on building all of its technology in-house, setting it apart from its competition. Additionally, it leverages telemetry and incident response tools like Sentry, Honeycomb, and AWS CloudWatch to monitor, centralize, and digest its data and overall infrastructure health.
However, navigating dozens of dashboards wasn't scalable for such a lean team. Only two engineers had the full-system context required to effectively troubleshoot complex incidents. The remaining engineers were required to make difficult tradeoffs:
“Do you fix the website, or do you stop to train someone on another tool? We couldn't afford both.”
Diego Lugo, VP of Engineering at Eaze, was constantly faced with this strategic dilemma. Engineers became narrowly expert in a couple tools, but could not effectively connect the dots across all the platforms, dashboards, and tools.
Moreover, and most importantly, some issues simply were not registering on any tool or on anyone's radar. Like an asymptomatic disease, how do you find a cure if you were never made aware of the ailment? In this case, Eaze had a silent issue that went undetected by all tools and all censors. There were no metrics to flag the issue because no one knew to monitor for it.
“The scary thing is: Who knows how long we would have gone before finding it?” pondered Lugo. Afterall, you cannot see what you were not looking for.
Once Eaze decided to onboard TierZero, the AI agents went to work immediately. It ingested Eaze's entire telemetry, deciphering information across all of its data sources (inclusive of its software applications and third party analytical tools), and producing a root cause analysis for the issues at hand.
In fact, promptly after deployment, TierZero began to flag anomalies it was detecting, urging the Eaze engineers to take a deeper look. It was connecting dots that engineers would not have thought to connect previously, and it began to sound the alarm on the aforementioned silent yet critical issue. TierZero had identified the undocumented process in Eaze's infrastructure which had been completely overseen.
According to Lugo, “We wouldn't have found it without TierZero. It wasn't revenue-impacting, so we weren't going to see it in dashboards. It wasn't latency, so it didn't show up in our performance monitors. It wasn't hurting customers, so no one complained. TierZero flagged it anyway. That's what's so incredible.”
Not only did TierZero identify the issue, but it also helped Eaze better understand the root cause of the undocumented infrastructure process that was critical to their software operations.
Previously, Eaze had to make strategic decisions regarding the bandwidth of its engineers. Now, with TierZero in place to monitor, flag, and tackle incidents, it frees engineers to take on the most pressing incidents head on.
"TierZero opens our eyes to things we could not have seen before. Even if we had seen an instance of the issue, without the resources to investigate the impact or scale of the problem, we might not have known to fix it!"— Diego Lugo, VP of Engineering at Eaze
For Eaze, TierZero is now the starting point for incident response and remediation. Whenever an issue is flagged, TierZero has the timeline, analysis, and root cause ready for review before the engineering team can even say "hello" on a remediation call.
Finally, with TierZero in place, Eaze's engineers have gained the confidence and skills to leverage this AI teammate for their overall advancement. They now have the time to learn and form insights from the other tools.
"...in the two months since onboarding TierZero, we've created more graphs and alerts that go into other systems than we have in the last year!"
Overall, TierZero fundamentally changes Eaze's approach to incident management. TierZero serves as a proactive, always-on infrastructure teammate, constantly identifying missing key insights to the overall business operations and prompting the engineering team to build sensors across all critical functions. TierZero didn't just solve a problem—it reshaped how Eaze's engineering team operates. AI isn't just a tool for Eaze anymore; it's part of the team.
"TierZero didn't just help us fix one issue. It opened the door to a completely new way of working. We're using more tools, adopting AI across the board, and finally have the breathing room to think strategically."
TierZero uncovered and helped resolve a critical bug that was undetected by all other tools. If left unchecked, it would have cost Eaze over $150,000; the damage may have been drastically worse the longer it went unnoticed.
In just two months, Eaze has created more alerts and monitoring graphs than in the entire previous year. With TierZero as the first line of investigation, engineers can begin with a root cause assessment and a powerful AI teammate that can answer system questions to tackle any threat.
Freed from the constant firefighting, engineers have gained agility and confidence. Engineers now view AI as an enabler, boosting morale and retention as a result.