Skip to content
Tenzok
Home
Services
Student Projects
BlogAboutContact
All student projects

4 scoped briefs

2 mini · 2 major

Test Automation & SDET

Most student QA projects are twenty recorded happy-path scripts that pass, prove nothing, and fall apart the first time CI runs them in parallel. This domain treats testing as engineering: cases aimed where the software is most likely to break, locators and waits that survive a UI change, coverage you can point at — a matrix naming every journey as tested or explicitly not — and flake handled deliberately instead of by retrying until green. A suite is worth exactly what it catches, so every project here ends with a defect log: reproducible steps and evidence, not a screenshot of a passing run.

PlaywrightTypeScriptSelenium 4TestNGREST AssuredPactk6promptfooGitHub Actions
Start a projectDiscuss a custom brief

The standard

Build the system. Understand every trade-off.

Every brief ends with working software, documented decisions, and evidence you can present—not a folder of code you cannot explain.

Two audiences, one engineering standard

Academic depth meets production discipline.

For students

Pick a brief below, or bring the problem statement your department handed you. We architect it with you, review every commit, deploy it to a real URL, and rehearse the viva until no question surprises you.

See student support

For companies

Test strategy and automation frameworks, turning a manual regression pass into a CI gate, stabilising suites the team has learned to ignore, API and contract coverage, load tests with thresholds that fail the build, and regression gates for LLM features someone else has already built.

Discuss a company build

Mini projects

Focused scope. Real engineering.

Three to four weeks. Narrow enough to finish, deep enough that you learn the decision that actually matters.

2 briefs
Mini project3–4 weeks

Playwright E2E Suite for a Deployed Web App

A cross-browser end-to-end suite over a live app with a page-object structure — the hard part is making it pass twenty runs in a row, not once.

PlaywrightTypeScriptGitHub ActionsAllure

You walk away with

  • Page Object Model built on role and test-id locators — no structural XPath that a DOM refactor breaks
  • Critical-journey coverage matrix: every journey listed, each one either mapped to a named test or marked untested
  • Cross-browser run (Chromium, Firefox, WebKit) green in GitHub Actions on every push
  • Twenty-consecutive-run stability report — every flake either fixed or quarantined with a written reason
  • Defect log with reproducible steps, a Playwright trace, and a screenshot per finding
Mini project4 weeks

Selenium + TestNG Regression Suite with BDD Specs

A Java regression suite a non-engineer can read and sign off: Gherkin scenarios, thread-safe parallel execution, and test data seeded over the API instead of clicked into existence.

Selenium 4JavaTestNGCucumberREST AssuredAllure

You walk away with

  • Gherkin feature files reviewed and signed off by someone who does not write code
  • Page Object Model with explicit waits — zero Thread.sleep anywhere in the repository
  • Parallel TestNG execution with a ThreadLocal WebDriver; a driver shared across threads interleaves commands and produces nondeterministic failures
  • REST Assured setup and teardown that seeds each scenario's data through the API and removes it afterwards, so no test depends on another test's leftovers
  • Allure report with per-step detail and screenshot-on-failure, plus a defect log naming the failing scenario with reproducible steps

Major projects

Capstones you can defend.

Ten to twelve weeks. Architecture, trade-offs, failure modes, deployment, and evidence—the project that carries an interview.

2 briefs
Major project10–12 weeks

Complete SDET Framework

One framework over an app you deploy yourself — UI, API and consumer-driven contract layers sharing a data factory, a sharded CI matrix, flake quarantine, and a k6 load test whose SLO thresholds fail the build.

PlaywrightTypeScriptAjvPactk6GitHub Actions

You walk away with

  • Three layers in one repo — E2E, API and contract — against an application you deploy and control (reuse the app from the Playwright mini), sharing one test-data factory that creates and tears down its own state
  • API layer validating every response against the OpenAPI schema with Ajv rather than hand-picked field assertions, plus negative and boundary cases: auth failures, malformed payloads, pagination edges, rate limits
  • Consumer-driven Pact contract published to a PactFlow free-tier broker (provisioning it is a week-one setup step) and verified in CI by the provider's own verification task — which you can run because you own the provider
  • Sharded GitHub Actions matrix with merged blob reports and a per-shard timing breakdown, plus flake quarantine: retry telemetry, a quarantine lane, and pass-rate-per-test trended in an off-the-shelf tool (Allure TestOps, or Grafana over a results table) rather than a dashboard you build yourself
  • k6 load test against your deployed environment with p95 latency and error-rate thresholds wired as a build gate, and a defect log with traces and — where identifiable — the commit that introduced each regression
Major project10–12 weeks

CI Regression Gate for an LLM Application

The app already exists — the RAG chatbot from AI & LLM Applications, or any deployed LLM app you own; building it is out of scope. This is the layer that domain does not ship: a versioned golden set, a calibrated judge, and a CI gate that blocks a prompt change that regresses quality.

PythonpytestpromptfooGitHub Actions

You walk away with

  • Versioned golden set of 150 prompts with reference outputs, labelled against a written rubric by you and a second annotator (a classmate), versioned in the repo alongside the prompts
  • LLM-as-judge scored against those labels and reported with its agreement rate against the human set — an uncalibrated judge measures nothing
  • Groundedness/hallucination and safety-refusal suites, each with a pass threshold over the whole set rather than a string match per case
  • Measured cost and latency budget per prompt version, with the API spend capped by design: a 20-prompt smoke subset on every push, the full set on pull request only
  • GitHub Actions gate that blocks any prompt or model change scoring below baseline, plus a defect log of the regressions it actually caught

Explore more

Other engineering domains

Search every brief
Python Full-StackJava & EnterpriseAI & LLM ApplicationsMachine LearningDeep LearningData EngineeringMobile AppsCloud & DevOpsCybersecurityIoT & EmbeddedJavaScript Full-StackVLSI & Chip DesignPower Electronics, EV & Energy SystemsSignal Processing & Wireless CommunicationsRobotics, Drones & Autonomous SystemsData Analytics & Business IntelligenceComputer Networks & Protocol Simulation
Your next build starts here

Turn the idea into software people trust.

Bring us a product brief, a business problem, or a final-year project. We’ll turn it into a clear scope, a working build, and a handover you fully own.

Start Your ProjectSend your brief

Prefer email? info@tenzok.in

Tenzok

A product engineering studio for ambitious companies, founders, and students who want real, production-minded work.

info@tenzok.in

Company

HomeBlogAboutContactFAQ

Services

MentorshipStudent ProjectsCompany ServicesDigital MarketingLaunch Support

Project domains

Python Full-StackJava & EnterpriseAI & LLM ApplicationsMachine LearningExplore all 18 domainsRSS feed

© 2026 Tenzok. All rights reserved.

Obsession · Purpose · Excellence