Lisna Thomas

About

Software QA engineer with 6+ years of experience in manual and automated testing across digital health, regulated gaming (web, iOS and Android) and client web applications.

Most recently at Altera Digital Health, where I was one of the AI Champions who built an AI agent workflow that turns requirements into BDD tests, runs them with Playwright and commits the code.

I verify features against design specs, Figma designs and acceptance criteria, manage the bug backlog in Jira and Azure DevOps, and read and write code in C#, Java, JavaScript and SQL. My focus is healthcare, where accurate and secure patient data is the measure of quality.

Domains I've tested

Healthcare (EHR, HL7, FHIR)
Regulated gaming (web, iOS and Android certification testing)
Client web applications

{
  "resourceType": "Practitioner",
  "id": "lisna-thomas",
  "active": true,
  "name": [
    {
      "use": "official",
      "text": "Lisna Thomas",
      "family": "Thomas",
      "given": ["Lisna"]
    }
  ],
  "telecom": [
    {
      "system": "email",
      "value": "lisnathomas99@gmail.com"
    },
    {
      "system": "url",
      "value": "https://github.com/lisnathomas"
    },
    {
      "system": "url",
      "value": "https://www.linkedin.com/in/lisna-thomas-137299261/"
    }
  ],
  "qualification": [
    {
      "code": {
        "text": "Bachelor of Computer Information Systems"
      },
      "period": { "start": "2018-09", "end": "2020-04" },
      "issuer": {
        "display": "University of the Fraser Valley"
      }
    },
    {
      "code": {
        "text": "Bachelor's degree in Computer Science"
      },
      "period": { "start": "2015", "end": "2018" },
      "issuer": { "display": "Mahatma Gandhi University" }
    }
  ]
}

My profile, as a healthcare system would see it.

Experience

My career as a CI pipeline. Select a stage to open its log.

QA Engineer, Altera Digital Health

Remote, Canada · EHR and interoperability software for hospitals and health systems (formerly part of Allscripts)

AI agents: As one of Altera's AI Champions, co-built an AI agent workflow (Claude) that reads requirements, generates Reqnroll feature files, runs the tests with Playwright, commits and pushes the test code, and produces a demo of the results
Demoed the AI agent workflow to business analysts, helping bring AI into the team's QA process; the workflow cut test-case writing time by 70% per story and reduced manual testing to about 20% of its previous effort through automation
Used Claude as an AI coding agent to generate test cases from user stories, write automation scripts and handle Git commits and pushes, integrated with Azure DevOps and Visual Studio
Verified clinical features against requirements and acceptance criteria, catching gaps between what was specified and what was built before release
QA for a 16-month feature integrating an external API for cross-network patient document search, with subscription create, cancel and notification workflows; owned test design, automation and end-to-end testing through release
Maintained automated UI tests with Selenium WebDriver, C#/.NET and Reqnroll (BDD), and automated desktop application tests with Appium, investigating failures and fixing flaky tests
Managed the bug backlog in Azure DevOps: reproduced, documented and tracked defects to resolution
Tested REST APIs with Postman and Swagger, validated data with SQL in SSMS, and validated HL7 messages and FHIR resources while protecting patient health information (PHI)

Results

Real numbers from real projects, reported the way a test run reports them.

Tests:: 6 passed, 6 total
Failed:: 0
Time:: 6+ years

Passed:70%less time writing test cases per story, with manual testing reduced to about 20% of its previous effortAI agent workflow · Altera Digital Health
Passed:16-monthinteroperability featureCross-network patient document search · Altera Digital Health
Passed:50%less manual testing effortSelenium framework · GLI
Passed:40%better build and deployment efficiencyCI/CD integration · GLI
Passed:25%more test coverageAPI automation with Postman · GLI
Passed:98%on-time releasesAgile and Waterfall projects · GLI

Projects

Each project opens the way a good test does: Given, When, Then.

OpenMRS AI Test Automation
@healthcare
Given
OpenMRS, an open-source electronic medical record system used by clinics worldwide
When
user stories are turned into BDD scenarios and run against the web UI and the FHIR API
Then
patient search, data consistency and access control are verified, with evidence saved for every run
- Built Playwright (C#) UI tests for the clinician login flow and FHIR R4 API tests for patient search, data consistency and access control, in one Reqnroll (BDD) suite
- Designed the suite for fast failure diagnosis: every run saves captioned videos, Playwright traces for step-by-step replay, and every API request and response, so a failing test can be investigated instead of blindly re-run
- Used user-facing locators and automatic waiting to keep tests stable, and debugged real environment and build issues across Docker, framework updates and file locks
- Built an AI bot (Claude API) that turns user stories into BDD scenarios using only existing steps, and checks every step before tests run
- Synthetic data only
Tech
- C#
- Playwright
- Reqnroll
- FHIR R4
- Docker
- Claude API
Links
- Code: OpenMRS AI Test Automation (opens in a new tab)
- Live test report: OpenMRS AI Test Automation (opens in a new tab)
School ERP AI Test Automation
Given
a school management web app with four user roles, built by Richu Thankachan (opens in a new tab)
When
each of the four roles logs in
Then
a Playwright login suite verifies the login for every role
- Playwright login suite covering all four user roles
- I built the test automation; the app itself is the work of a developer collaborator
Tech
- Playwright
Links
- Code: School ERP AI Test Automation (opens in a new tab)
- Live test report: School ERP AI Test Automation (opens in a new tab)

How I use AI in testing

At Altera Digital Health, I was one of the AI Champions who co-built an AI agent workflow that reads requirements, generates Gherkin feature files, runs them with Playwright and commits the code, cutting test-case writing time by 70% per story. AI helps me test faster. These four rules keep it honest.

AI drafts, the framework constrains
The bot may only use steps that already exist.
Every AI output is checked before anything runs.
Humans own the risk decisions.
Synthetic data only, never real patient data.

Skills

Digital health

HL7
FHIR R4
EHR systems
Healthcare interoperability
Patient data privacy (PHI)

Test automation

Playwright
Selenium WebDriver
Appium (desktop applications)
Reqnroll (SpecFlow successor)
BDD/Gherkin
Katalon Studio
NUnit
TestNG
JUnit
Page Object Model
End-to-end (E2E) testing

AI in testing

AI agents
Agentic test automation workflows
Claude
Claude API
AI-generated test cases
AI-driven requirement review
Prompt engineering
Validating AI output

Programming

C#
.NET
Java
JavaScript
React Native
SQL

Bug tracking and specs

Jira
Azure DevOps (Boards, Test Plans)
Figma
Design specs
Acceptance criteria
Bug backlog management
Defect triage
Clear bug reports with evidence

CI/CD and tools

CI/CD pipelines
Git
GitHub
Docker
Visual Studio
Splunk

API and data

REST and SOAP APIs
Postman
Swagger
SoapUI
SQL Server (SSMS)
JSON

Testing types

Functional
Regression
Smoke
Sanity
Integration
End-to-end
Exploratory
Edge-case
Cross-browser
iOS and Android mobile
Desktop
API
Database
UAT
Release

Methodologies

Agile/Scrum
Waterfall
SDLC
STLC

Contact

Open a ticket. It drafts an email in your own mail app, so nothing is sent or stored by this site.

About

Domains I've tested

Experience

QA Analyst, Concentrix

Test Engineer, Gaming Laboratories International (GLI)

QA Engineer, Altera Digital Health

AI-assisted test automation, Portfolio projects

Results

Projects

OpenMRS AI Test Automation

Tech

Links

School ERP AI Test Automation