Release With Confidence. Know What Needs Fixing First.
Software testing for web and mobile teams preparing to launch, improving release quality, or adding QA capacity. Work with Carnelian's team across Dubai and India for exploratory testing, automation and ongoing QA support.
Where Teams Usually Start
Pick the situation that sounds like yours
"We are launching soon"
Release readiness assessment
- What we need from you
- A staging build, test accounts, your release date and the journeys that matter most.
- What you receive
- Tested critical journeys, every defect with steps and evidence, and a list of remaining risks.
- How we measure success
- You launch knowing what is broken, what is fixed and what risk is left.
"Regression takes too long"
Automation assessment and build
- What we need from you
- Access to a test environment, your current manual test cases and how you release today.
- What you receive
- A prioritised coverage plan, maintainable automated tests and execution reports in your pipeline.
- How we measure success
- Regression checks run with every release, and manual effort shrinks release by release.
"Our developers need QA support"
Dedicated QA engineer or team
- What we need from you
- Access to your issue tracker and chat, your sprint rhythm and a product walkthrough.
- What you receive
- Agreed QA capacity inside your workflow, weekly reporting and a QA lead overseeing quality.
- How we measure success
- Fewer defects reach production and developers spend less of their time testing.
"Users report problems we cannot reproduce"
Exploratory and compatibility testing
- What we need from you
- The reports you have, any analytics, and the devices and browsers your users rely on.
- What you receive
- Reproducible findings across agreed devices and environments, with screenshots, recordings and logs.
- How we measure success
- Vague complaints become reproducible bugs your developers can fix.
What Late Bugs Cost
Industry estimates, not our own results. They all point the same way: the earlier a bug is found, the less it costs to fix.
$2.41 trillion
Estimated cost of poor software quality in the US in 2022, at a minimum.
Source: CISQ, Cost of Poor Software Quality in the U.S.: A 2022 Report (opens in a new tab)$59.5 billion a year
Cost to the US economy of software bugs that are not found and fixed as they appear.
Source: NIST Planning Report 02-3, 2002 (opens in a new tab)$22.2 billion a year
Saving the same study estimated from realistic improvements in finding and fixing bugs earlier.
Source: NIST Planning Report 02-3, 2002 (opens in a new tab)See How We Work
Real reports and real investigations, not promises
Sample Report
Eleven real functional findings from HiNeighbour, our live app, each with steps to reproduce, root cause, fix and retest.
Read the sample reportCase Studies
A five year dedicated QA team, system testing for a video platform, sign in and sync for a food diary app and more, all from public Upwork work.
Read the case studiesOur Own Apps
HiNeighbour and ABS Twin run in production on the App Store and Google Play, tested by the same team that tests yours.
See our productsWhat Clients Say
Word for word from public reviews on Upwork
"She found multiple bugs in my application in no time."
"Her reports were detailed and extremely useful."
"He spent countless hours during the weekend testing the latest code changes."
Who Leads Delivery
Sayed Shaista, Founder and QA Lead
Fifteen years of QA across healthcare, fintech and SaaS, including two years testing AI products for startups. Has interviewed, hired and mentored testers, and oversees Carnelian's QA engagements and audit reports.
On Upwork: Top Rated Plus, 100% Job Success, rated 5.0. See the profile(opens in a new tab)
Connect on LinkedIn(opens in a new tab)Ways to Work With Us
App Audit
Fixed US$399 for one website or app. Report within three working days of scope and access, plus one retest.
Explore the App AuditRelease Readiness Project
A fixed scope test cycle before a launch or major release, quoted after a short call.
Request a QuoteDedicated QA Team
One engineer or a full team working in your tools, scaled month to month.
Explore Dedicated QAAutomation Build
We design and build your automated regression suite, then maintain it if you want us to.
Explore AutomationHow an Engagement Runs
01
Scope
We agree the product, journeys, devices and timeline in writing.
02
Plan
Test charters and a coverage plan, shared with you before testing starts.
03
Test
Exploratory sessions and scripted checks, logged as we go.
04
Report
Every defect with steps, evidence, severity and business impact, plus a walkthrough.
05
Retest
We verify your fixes and update the report.
Every engagement delivers
- Bug reports with steps, severity and business impact
- Session charter reports from exploratory testing
- Screenshots, screen recordings and logs as evidence
- A summary of remaining risks before release
What AI Changes in Practice
We use AI to draft test ideas from requirements, generate varied test data and sort large crash and log dumps. A tester reviews everything before it reaches your report, and AI never decides whether something passes. When your product has AI features of its own, we test their behaviour directly: wrong answers, prompt injection, data leakage and how they fail.
Explore AI and automation testingQuestions, Answered
Do you test on real devices?
Yes. Manual testing runs on real iPhones, Android phones and desktop browsers, not only emulators.
Can you work inside our tools and sprints?
Yes. Our engineers work in your issue tracker and chat, join your stand ups and follow your release rhythm.
Which tools do you automate with?
Playwright, Cypress, Selenium and Appium for functional automation, and JMeter for load testing. We choose with you, based on your stack and who will maintain the tests.
Where is the team?
Our headquarters is in Dubai, with a delivery centre in India.
Do you sign an NDA?
Yes, before we see anything.
Tell Us What You Are Releasing Next
We reply within one working day.