Kostiantyn Shkliar builds test automation for systems where getting it wrong has real consequences.
He has spent over ten years in automation architecture and software quality engineering (SDET), focusing on fintech and large enterprise CRM systems. In recent years, he has worked in the U.S. financial services market as a Senior Salesforce QA Automation Engineer, supporting platforms tied to $130B+ in assets under management.
A key result from his work is a measurable change in how teams validate releases: using a hybrid validation model, he helped reduce regression execution time from about 120 hours to about 2 hours. That shift mattered because it reduced release friction while keeping confidence in the system’s most important requirements: security behavior and data integrity.
This is a biographical look at how his career moved from setting up automation processes in large organizations to building validation systems that stay stable as Salesforce environments scale.
Kostiantyn Shkliar’s Background in High-Stakes QA Automation
Kostiantyn started at EPAM Systems, where he helped set up QA and automation processes for global banks and large retail chains. Early work in large environments shaped his habits: focus on repeatability, clear validation signals, and automation that can be maintained over time.
As his career progressed, he stayed close to systems where quality connects directly to trust and risk. That led him deeper into financial services and into Salesforce-based enterprise CRM, where configuration changes can affect access boundaries and data correctness across many users.
Why Regression Testing Turns Into a Release Bottleneck
In mature enterprise systems, complexity tends to build gradually. Rules grow. Integrations expand. User roles multiply. The test suite grows too, often by adding coverage whenever something breaks.
Over time, regression testing becomes the slowest step in releasing. That brings predictable issues:
- Feedback arrives late
- Issues are discovered after more work has already been built on top
- Teams feel pressure to reduce validation to keep delivery moving
Kostiantyn’s work has focused on shortening regression cycles without weakening confidence in results.
The 120-Hour Regression Cycle and the 2-Hour Outcome
At a large U.S. financial company, Kostiantyn designed and implemented a hybrid validation model that changed how regression testing ran.
Regression execution dropped from 120 hours to 2 hours. That shift supported a move toward daily releases while maintaining confidence in data integrity. In practical terms, it meant engineers could get answers the same day a change was made, and the business could plan releases around evidence rather than long waits.
Hybrid Validation, Explained Without Jargon
Hybrid validation combines different types of checks based on what needs to be proven. The goal is simple: use the most reliable signal for each kind of risk.
1. Targeted UI automation
UI tests cover a small set of critical end-to-end flows. They confirm key user journeys without carrying the full burden of regression validation.
2. API-level and metadata-driven validation
Much of the validation runs through APIs and metadata to verify business rules and configuration outcomes in a stable way. In Salesforce environments, this supports validation of security configuration behavior, including Permission Sets.
3. Data integrity checks with controlled test data
Validation includes confirming that data remains consistent through updates and complex operations. Kostiantyn designs scalable test data generation systems (“Data Factories”) using JSON schemas so datasets can be created predictably and reused.
Together, these layers reduce noise and keep the suite fast and dependable.
Why UI-Only Automation Breaks Down in Large Salesforce Orgs
UI automation is easy to start with, but at scale it becomes sensitive to timing and interface changes. Tests fail for reasons that don’t reflect real risk, which increases false positives and time spent investigating noise.
When teams spend too much time chasing noise, trust in the suite drops. Kostiantyn keeps UI automation focused and uses it as one layer, not the foundation. This is also where the hybrid model helps: even when the UI changes, core validation can still run against stable interfaces.
Architectural-Driven Validation: Focus on Permissions, Logic, and Data
Kostiantyn describes his approach as architectural-driven validation: verify the parts of the system that define correctness, not only what appears on the surface.
That means validating:
- Security models and access boundaries, including Permission Sets
- Business logic
- System behavior through APIs and metadata
- Data integrity during updates and complex operations
This approach matters most in environments where configuration changes are frequent. Small permission adjustments, rule updates, or new automation can create unexpected side effects, so validation has to stay close to how the platform actually enforces rules and access.
Core Skills: Salesforce Security, APIs, .NET, and Test Data Systems
Kostiantyn’s core skills include:
- .NET 8.0 and C#, with deep API-level interactions
- Salesforce architecture expertise, including the security model, Permission Sets, and LWC components
- Scalable Data Factories using JSON schemas
- SDLC optimization to reduce technical debt
Education: IT Foundations and Business Perspective
Kostiantyn holds a Master’s degree in Information Technology from Kyiv Polytechnic Institute (KPI). He also holds a second degree in marketing, which supports a business-focused view of engineering decisions. That mix helps him connect validation work to outcomes leaders care about, like release reliability, rework, and long-term maintenance cost.
Current Role: Keeping a Financial Services CRM Stable Through Change
Currently, Kostiantyn works as a Senior Salesforce QA Automation Engineer. His responsibilities include automation architecture, security configuration validation, and automation around complex financial operations, with the goal of minimizing release risk for a mission-critical CRM platform.
He maintains strict NDA compliance and does not disclose internal program names, specific business process details, or internal policies. He presents himself as an independent expert engineer and uses only publicly safe context, such as AUM scale.
How He Thinks About Quality: Trust, Evidence, and Simplicity
Kostiantyn treats automation as repeatable proof that a system stayed correct through change. His principle is:
“Quality is not something you add at the end. It’s what you build the system on from day one.”
He is inspired by Robert C. Martin’s Clean Architecture and the belief that complexity increases reliability risk. He aims to keep validation systems understandable and maintainable so teams can rely on them over time, even as the platform grows.
What He’s Working Toward: Hybrid Validation Research and AI for Security Weakness Detection
Kostiantyn plans to continue research in hybrid validation approaches and introduce AI-driven elements that can identify weaknesses in Salesforce security configurations. Long-term, he aims to work as a consultant focused on architectural reliability for fintech systems, helping teams keep release speed aligned with strong validation. n
:::info
This article is published under HackerNoon’s Business Blogging program.
:::
