Customer Trust Begins with Rigorous Validation
What do you currently do at AIO?
The Application Engineering Team analyzes issues that arise during customers’ product design-in and evaluation processes, including device recognition failures, performance degradation, data errors, and Firmware-related issues. We reproduce the customer’s Host environment and test conditions internally and use Logs and Protocol Traces to identify whether the root cause lies in Firmware, NAND, the Host, or power conditions.
We share the analysis results with the development team to determine the appropriate direction for Firmware modifications and support the entire process, from validating the changes through to customer re-evaluation. We also develop Log Parsers and automation tools to improve the efficiency of recurring analysis tasks.
What skills or capabilities do you consider most important in your role?
I believe the most important capability in AE is the ability to turn a complex failure symptom into a reproducible technical problem.
Based on the symptoms reported by a customer alone, it is often difficult to immediately determine whether the cause lies in Firmware, NAND, the Host Driver, power, Signal, or the test scenario. We therefore define the differences between normal and failure conditions, formulate hypotheses based on the Command Sequence, Timing, power history, and Logs, and then verify them step by step.
This requires a solid understanding of SD and eMMC Protocols, NAND, and Firmware architecture, as well as the ability to use C, Python, and Linux-based analysis tools. Equally important is the ability to clearly communicate the analysis process and the technical basis for our conclusions to both customers and the development team.
What has been your most rewarding experience?
One of my most rewarding experiences was resolving an intermittent device response failure that occurred under specific Power Cycle and re-initialization conditions during a customer Qualification process.
The initial Logs were not sufficient to clearly identify the Root Cause, so I reviewed the customer’s Command Sequence and power conditions and then conducted a step-by-step comparison of normal Logs, failure Logs, Protocol Traces, and the internal Firmware state.
Through this analysis, we identified a condition in a specific Initialization Flow where the Firmware state was not being restored correctly, which allowed us to determine the required modifications together with the development team. We then carried out internal reproduction testing and customer re-evaluation using the revised Firmware and confirmed that the issue did not recur. As a result, we were able to successfully complete a Qualification process that had been at risk of being suspended.
What would you like to achieve at AIO?
My goal at AIO is to evolve AE from a reactive function that responds after problems occur into a Predict & Prevent system that anticipates and prevents issues before they arise. I would like to organize the customer issues and Logs accumulated to date by type and build a Signature Database covering issues such as device recognition failures, performance degradation, and data errors.
I also hope to build an AE Platform that automatically collects the necessary information and performs an initial analysis when a similar issue is reported. By integrating Jira, Confluence, Log Parsers, and analysis tools into a single Workflow, the goal is to reduce analysis time and create an environment where engineers can focus more on Root Cause identification and problem solving. In the long term, I would also like to incorporate customers’ actual Workloads and requirements into product specifications and Firmware improvements.
What advice would you give to those preparing to join AIO?
If you are preparing for an AE role at AIO, I believe it is more important to develop the habit of tracing a problem to its root cause and verifying it thoroughly than simply trying to find the correct answer quickly.
Because problems in storage and Embedded Systems are rarely solved through Firmware alone, a broad understanding of C, Operating Systems, Device Drivers, Interface Protocols, NAND, and power behavior is highly valuable. Hands-on experience is also helpful—for example, collecting Logs from actual devices, setting up reproduction environments, or gaining experience in analysis and automation using Python and Linux.
Most importantly, the role requires not only finding technically accurate answers, but also accurately understanding customer requirements and translating the findings into actionable information for the development team. If you are comfortable approaching unfamiliar problems without fear and collaborating with colleagues based on evidence, I believe you can grow quickly at AIO.