AI Test Lead - W2 only

1. Has experience with migrating existing Selenium Automation suite to Playwright framework using Claude. When asked which Claude model did you use he was not sure first mentioned Haiku then said whatever latest model is in market.
2. Don't look to have experience with migrating legacy On-prem applications to new Tech Stack, where you don't have any proper documentation available.
3. He has experience with migration of web based applications.
4. We didn't hear agentic use for extracting requirements from legacy systems and building test framework. He was more inclined on human touch and going through UI application and building requirements.
5. He told about using AWS Bedrock OCR automation tool to read through application and generating tests. We asked how will you do for a desktop app, so he didn't had answer for that. So he will be figuring out a solution.
6. I didn't see an agentic approach to Automation framework development.
7. He does not have experience building a test framework using AI.


NOTES:
Project Overview: TechEvo Application Modernization

The Pitch: We are shifting our strategy from massive, slow application re-architectures to a highly targeted, AI-driven platform roadmap. Because bad actors are rapidly scaling AI models to find vulnerabilities, we need to stay ahead of the threat curve and "stop the bleeding." We are building internal AI agents to automate the modernization and security upgrades of our legacy applications.

The Technical Scope
The team is focused on two primary modernization tracks, utilizing Claude Code and Java:
Track 1: Infrastructure & Core Tech Upgrades
Migrating legacy applications running on older Linux platforms and older Java runtimes (e.g., upgrading from JDK 17 to JDK 25).
Building AI agents that can scan and understand legacy applications, generate technical specifications, and automatically execute the dependency-by-dependency upgrades.

Track 2: Security & Identity Modernization
Transitioning applications from legacy security/authentication models (LDP, ISD) to modern Okta integration.
Evaluating applications on a case-by-case basis to determine if they need a tech stack upgrade, a security upgrade, or both.
The Role: Agentic Engineers
We are not looking for standard developers to manually rewrite code. We are looking for Agentic Engineers.

The Core Task: Write, instruct, and constantly refine AI agents/skills to handle the heavy lifting of modernizing our infrastructure stack.

The Goal: Engineer the AI agents to successfully complete 90% of the upgrade work autonomously, leaving the remaining 10% for human verification and validation.

The Ideal Candidate Profile
Systems Thinkers: Must understand how massive enterprise systems interconnect. If they upgrade a database to a new server, they need to instantly understand the downstream impacts on connected systems.
The "Fixers," Not The "Finders": We do not need auditors who simply identify vulnerabilities. We need hands-on engineers who build the automated solutions to fix them.
Battle-Scarred Veterans: We are looking for engineers with real-world scars. A standard Senior Developer with 7 years of isolated coding experience will not survive here; they need a history of navigating complex, high-stakes system failures (P1 incidents).

Quick Screening Questions for the Recruiter
"Tell me about a time you had to modernize a legacy application. How did you map out the downstream impacts before making changes?" (Listen for systems thinking and architectural awareness, not just code-level changes).
"How many P1 (Priority 1) incidents have you been involved in during your career, and what did they teach you about system design?" (Listen for battle scars and lessons learned).
"If I asked you to automate a massive JDK upgrade across dozens of legacy applications today, how would you approach building an AI agent to do 90% of the work?" (Listen for an understanding of agentic engineering, automation, and prompt/skill design).