QA Manager

Yaseen Farooqui

I lead quality assurance and AI strategy for a non-standard auto insurance technology platform built on Guidewire InsuranceNow, covering more than ten states across personal and commercial lines along with the internal tooling and processes that support them.

QA review cycle reduction
0%
Triage response time reduction
0%+
Annual hours reallocated
0+
Annual savings delivered
$0K+

About

A coaching-first leader

I am a QA leader who understands testing strategy and am using AI to accelerate delivery while not compromising on quality.

Today

My responsibilities cover QA across a multi-state non-standard auto insurance technology platform, and the AI strategy that operates alongside it. The portfolio spans rollouts in 10+ US states, executive reporting tools used in board meetings, and an AI triage program that has reallocated more than 1,200 hours of annual labor away from manual ticket handling.

Operating Style

Senior IT leadership are my standing partners on AI strategy, SLA frameworks, and the dashboards that make IT impact easy to read at the executive level. A psychology background shapes a coaching-first approach: I invest in the people doing the work, and I invest in the tools that let leadership see what those people are accomplishing.

On my desk

Designing smarter AI-generated test cases. The goal is coverage that maps to real scenarios and the actual steps a tester would take through a complex system.

How I Work

Operating model

Measuring QA

Read more

I measure QA against escape rate and catch rate, with rework cost translated into dollars. Throughput metrics like tickets touched or test cases written tell leadership how busy a team is. What they do not show is whether the team is delivering quality. Escape rate is what the customer actually experiences after release. Catch rate is what QA actually prevented before it. First-time-right belongs in a different category. It measures developer submission health, and treating it as a QA performance metric punishes the team for upstream behavior. When the metrics are correctly separated, the conversation about QA investment moves out of headcount defense and into portfolio strategy. My team compressed the QA review cycle from seven workdays to two over the past year under this framework, and coverage expanded over the same period.

Executive reporting

Read more

I treat dashboards as decision tools rather than status reports. The reporting I build auto-derives status from live work-tracking data, so the executive view and the team view never disagree. The same pipeline produces a slide for the leadership review and a Teams card for the morning standup. The three executive reporting tools I built run on this principle and are now standing artifacts in leadership reviews.

Scoping AI initiatives

Read more

I scope AI work the same way I scope any infrastructure project. The model is the last decision, not the first. The triage program that now serves our IT organization began as a survey of every ticket category we received and what each one required to resolve. Only after that did the priority taxonomy and the intake form get designed, and only after those was a model selected. The system has held up across model upgrades because it was built around the workflow rather than around the model. The same discipline produced the internal AI tooling that supports our QA analysts during ticket review.

Coaching analysts

Read more

I lead with coaching because most QA performance issues are clarity issues, not capability issues. Analysts perform when expectations are explicit and the work queue is visible. The published policy handbook tells the team what good looks like, and the cadence I run includes weekly one-on-ones and a monthly QA Town Hall. Performance reviews are a record of conversations that have already happened rather than a surprise.

Resume

Experience

Jan 2025
Present

PRIS Corp (Warrior Invictus)

QA Manager and Test Automation Architect, Guidewire InsuranceNow

Greater Chicago Area (Hybrid)

Leadership & Reporting

My responsibilities cover quality assurance and AI strategy across a non-standard auto insurance technology platform built on Guidewire InsuranceNow, with rollouts active across more than ten states. Senior IT leadership receives my reporting on portfolio health and on the SLA framework that governs our AI programs.

Under my team the average QA review cycle compressed from 7.3 workdays to 2.0 workdays over the past year, a 72 percent reduction validated against EazyBI reporting. The metrics driving the compression were catch rate and escape rate, with rework cost translated into dollars rather than tracked as a count, and coverage expanded over the same period. New hires are brought up to platform certification and fully operational within sixty days under the onboarding program I established.

AI Programs

The AI-assisted support triage system I designed and shipped now fronts the IT helpdesk for the company's internal employees and external producer network. First response moved from regular SLA breach to under two minutes. SLA compliance moved to near 100 percent, and the program reallocates more than 1,200 hours of annual labor away from manual ticket handling.

Alongside the system, the supporting SLA framework I authored introduced a six-tier priority taxonomy and a redesigned intake form. The form lets the system derive priority from structured impact questions rather than from a user-selected dropdown, and the framework is reviewed quarterly as the operating contract for the program.

Internal AI tooling now in development supports QA analysts through Jira ticket evaluation, acceptance criteria gap analysis, and test scenario suggestion, with automated test script generation planned for a second phase. The system uses few-shot prompting and retrieval over our own Jira history and test repositories, which is what makes it context-aware rather than generic. The tooling is built around the existing review workflow rather than around the underlying model, so it has held up across model upgrades and remains in daily use.

Executive Reporting

Three executive reporting tools I built run on live work-tracking data: a QA KPI tracker, a state rollout readiness console, and an IT portfolio big board covering more than 100 initiatives. Each auto-derives status, so the executive view and the team view never disagree. All three are now standing artifacts in executive leadership reviews and have been transitioned to a successor for ongoing ownership.

Cost, Vendor & Process

Documented annual savings during the review period totaled more than $80,000. The largest contributor at over $60,000 was the AI triage program. Deployment smoke test automation contributed more than $20,000 by compressing release testing from fifteen QA hours to under one hour, and a vendor renegotiation produced roughly $2,300 while doubling licensed seats and increasing monthly cloud credits by 50 percent.

QA-led pre-refinement sessions, which I introduced, strengthen acceptance criteria before development begins and surface testing scope early. Kickback rates dropped by approximately 35 percent under the practice, and sprint predictability improved across the QA function.

Standards, Tooling & Community

The QA Policy Handbook, the Xray Test Case Management Guide, and the new-hire QA Analyst training program for the InsuranceNow platform are mine, including its integrations with comparative raters, payment processors, and adjacent policy administration systems. Xray adoption fell to me as well, from workflow design and evidence collection standards through Jira integration and team rollout.

A Rules Engine API test suite covering 129 active validation rules across the non-standard auto product line came out of this period. The suite runs through a Python execution harness against a Postman collection, with an automated reporting pipeline producing QA-ready output.

Monthly QA Town Hall meetings, which I founded and run, align quality practice across the IT organization. The AI Native certification rounds out the period.

Earlier roles

  • International (formerly Navistar)

    Tosca Architect and Technical Administrator

    Nov 2023Jan 2025 · Lisle, IL (Hybrid)

    Read more

    Leadership & Team Management

    Day-to-day leadership of the offshore QE team based in India was mine to run, coordinating across a 10.5-hour time zone difference between Central US and Indian Standard Time through structured early morning syncs and asynchronous review processes. The cadence I established covered code and automation script reviews along with the standards the team executed against.

    More than 40 QA resources came onto the platform under my onboarding program, and eight teams transitioned to Tosca within six months. Each team received reusable component libraries built to the standards I authored, along with training delivered directly by me and post-launch support during stabilization.

    Technical Architecture & Implementation

    The enterprise Tricentis Tosca implementation across the IT organization fell to me to own. The work included license server architecture, repository segregation across development, QA, and production environments, Azure Active Directory single sign-on integration, and qTest configuration for test management.

    A CI/CD-integrated regression suite executing against Azure Pipelines came out of this work, which gave the organization triggered regression capacity for the first time. The regression footprint extended into IBM z/OS mainframe via the TN3270 module, including CICS transaction validation, bringing a layer of the application stack under automated coverage that had been entirely manual. Coverage expanded further into SAP and into the API layer.

    NeoLoad came in for performance testing alongside the test data management practice I authored, including the data obfuscation processes required to support automation while maintaining data security compliance.

    qTest came online for test and defect management across three teams under my configuration, implementing the Manager, Launch, and Insights modules and the workflows that connected them to the Tosca execution layer. Executive dashboards for Associate Directors covered automation progress and test coverage trends, and Tricentis licensing was negotiated directly with the vendor to align license counts and server allocations to actual enterprise demand.

  • TrimedX

    Senior Test Automation Engineer, Healthcare Asset Management

    Jan 2022Nov 2023 · Indianapolis, IN (Remote)

    Read more

    Build-out of the Tricentis Tosca automation footprint for a healthcare asset management platform fell to me. The platform supported equipment tracking and compliance workflows in clinical environments, and the work covered UI, API, and database validation in a single architecture, which let one regression run produce evidence across the full data path rather than only the user interface.

    A regression suite of more than 250 test cases reached approximately 80 percent functional coverage of the platform. The reusable test library I wrote decoupled test logic from application selectors and from test data, cutting automation maintenance effort by roughly 40 percent and becoming the basis for the team's training material on maintenance practice.

    Defect triage sessions and root cause analysis across the QA team ran under my leadership, alongside the internal standards I authored for how defects were classified and routed back to development. Test plans and test strategy documents I created aligned QA execution to the business objectives the product owners had defined, and I participated in SAFe ceremonies including sprint planning and Program Increment planning.

  • State of Michigan

    Test Automation Engineer, Government Web and Mobile

    Dec 2020Jan 2022 · Lansing, MI (Remote)

    Read more

    Web automation in Cypress and mobile automation in Appium across iOS and Android device matrices came out of this period, with API integration coverage for the public-facing applications I supported. Database validation ran directly against the state systems of record where applicable.

    Accessibility compliance testing under the Americans with Disabilities Act used JAWS for screen reader validation on Windows, VoiceOver for iOS and macOS, and TalkBack for Android. The work covered both functional accessibility validation and the audit documentation required to demonstrate compliance.

    Test infrastructure for cross-device mobile validation came out of my work as well, including device matrix selection criteria and the test data management practices required to run the same scenario consistently across platforms with materially different operating system behavior. Integration testing across the public-facing applications covered authentication flows and data exchange with state systems of record, including validation of citizen-facing forms that fed downstream case management workflows.

  • Nationwide Insurance

    Quality Assurance Engineer and Tosca Engineer

    Jul 2018Dec 2020 · Brea, CA (Remote)

    Read more

    More than 100 automated test scenarios and 500+ test cases contributed to the enterprise regression suite came out of this period, all supporting Nationwide's property and casualty applications. Reusable automation infrastructure with module and page object patterns in Tricentis Tosca came out of this work, alongside contributions to the enterprise regression standards through the component library and the shared test data management practices that supported reuse across teams.

    Load testing in JMeter ran against transaction-critical endpoints, including the test data and execution profiles for sustained load and peak load scenarios. API testing in Postman covered authentication and data validation, with contract verification against downstream consumers.

  • Apple

    Technical Specialist and Genius Bar

    Jun 2014Jul 2018 · Various Locations

    Read more

    Hardware and software diagnostics, customer education, and de-escalation in a high-volume retail environment defined the role. Composure under pressure and systematic troubleshooting on iOS and Mac platforms using internal diagnostic tools were the daily work, with regular translation of technical concepts to non-technical customers.

    Component-level repairs covered display replacement, RAM installation, and storage replacement, plus data migrations between devices using internal tooling while maintaining data integrity and customer privacy. Recurring service patterns surfaced through this work informed structured feedback to leadership that drove local process changes.

Selected work

Programs & writing

AI-Powered Support Triage

An n8n-based automated intake and triage system fronting a major helpdesk platform. First response now lands in under two minutes, and the supporting SLA framework was presented to senior leadership at multiple executive reviews.

Triage first response moved from regular SLA breach to under 2 minutes; 1,200+ hours of annual labor reallocated to ticket resolution.

n8nAICIO Reviews

IT Portfolio Big Board

A department-wide portfolio tracker covering 100+ IT initiatives across multiple teams. One Python pipeline produces a board-ready PPTX, an Excel Gantt, an interactive HTML dashboard, and a Teams card. The whole thing is config-driven, so non-technical stakeholders can edit it live during a meeting.

Reduced executive readout assembly from a multi-day analyst pull to a single rebuild step.

PythonPPTX / Excel / HTMLBoard Reviews

Go-Live Readiness Console

A real-time readiness console for a multi-state insurance platform rollout. It auto-derives epic health, models team capacity, and ships a self-contained HTML dashboard with a Claude-powered Q&A agent that answers questions over live program data.

Go / no-go decisions now made from a single live readout instead of a multi-day data pull.

Node.jsClaude APIExecutive Reviews

QA KPI Framework

A Node.js system that pulls live work-tracking data and produces an Excel workbook, an interactive dashboard, and a daily Teams card. It defines the metric framework (catch rate, escape rate, first-time-right) and translates rework into dollars, so QA investment can be discussed with leadership in concrete terms.

Reframed QA investment conversations in dollars rather than percentages; metrics now feed both sprint review and 1:1 coaching.

Node.jsJira APISprint Review + 1:1

Skills

What I work with

Leadership
QA Strategy & SLA FrameworksExecutive & Board ReportingAI Strategy & GovernanceVendor NegotiationCoaching & Performance ManagementCross-Functional Partnership
AI & Automation
Anthropic Claude (Code, Desktop, MCP)GPT-4o / OpenAICustom AI Agentsn8n Workflow AutomationVoiceflowAtlassian MCPmabl MCP
Test Automation
mablTricentis Tosca (AS1/AS2/TDS/API/SAP)CypressSeleniumAppiumJMeterPostmanDatadog Synthetics
Engineering
Node.jsPythonJavaScriptSQL (MSSQL, Oracle)Microsoft Teams Adaptive CardsExcel / PPTX / DOCX automationGit / GitHub
Project & Test Management
JiraXrayqTestAzure DevOpsEazyBIAgile / Scrum / SAFe
Domain
Guidewire InsuranceNowMulti-state non-standard auto rolloutsComparative raters (ITC, EZLynx, Vertafore, Quomation)FNOL, IVR, and payment integrations
Certifications
AI NativeISTQBTosca AS1 / AS2Tosca TDS1 / TDS2Tosca AS1 for SAPTosca API Specialistmabl Certified

Contact

Get in touch

The best way to reach me is by email or LinkedIn.