Ray Rahat

• Designed an e2e Automation Framework using Playwright and Javscript for UI and API; as well as LangGraph ReAct agent (LangChain + Playwright MCP) that reasons over tools: read the ticket, explore the page, generate Page Object Model and Playwright specs, run them, and repair until they pass.
• Built an AI Agent validation pipeline that extracts data from Excel with an LLM, then reconciles it field-by-field against a live GraphQL API using the latest intake plan — so spreadsheet errors show up without a manual review.
• The same agent has an exploratory mode: given only a URL, it drives a live browser, drafts a test plan plus happy-path and negative suites, executes the cases, and writes a pass/fail defect report — no ticket and no generated automation code.

• LangGraph owns the loop (which tool to call next, when to retry, when to stop); Python verification then re-runs the Playwright spec independently so the agent is never trusted on its own claim of success.

AI Systems & Automation Engineering

IDP -Large Intelligent Document Processing

• Automated the triage process for ~6,000 monthly IT support tickets using an AI-driven classification pipeline, significantly reducing manual QA and support workload.
• Extracted raw ticket data from Google BigQuery, performed preprocessing, and applied OpenAI LLMs (via LangChain and LangGraph) for summarization and classification — distinguishing between user errors and system bugs.
• Developed structured evaluation methods for LLM outputs to assess factual accuracy, coherence, hallucination rates, and user intent alignment.
• Applied BLEU, ROUGE, and embedding similarity scores to benchmark the quality of summarization and classification.
• Created prompt robustness tests using edge-case and adversarial inputs to evaluate LLM resilience and reliability.
• Implemented automated testing for prompt injection vulnerabilities and unexpected completions using adversarial prompt suites.
• Designed a reusable prompt validation pipeline using LangChain to ensure prompt safety, stability, and clarity across deployments.
• Built a model-in-the-loop evaluation framework using OpenAI to self-assess output consistency and flag bias, toxicity, or incoherence.
• Enabled LLM-to-LLM review workflows, where one model critiques another’s output to expose nuanced flaws and hallucinations.
• Integrated output metrics with Datadog dashboards to track model behavior and detect performance drift over time.
• Ensured alignment with internal AI governance and ethical AI guidelines by incorporating safety guardrails and documenting ambiguous or unsafe outputs.
• Automatically created Jira tickets for confirmed bugs, enabling faster dev triage and feedback loops.
• Built the entire system in Python, with orchestration via LangChain and LangGraph, and integrated pipelines into CI/CD for continuous model evaluation and monitoring.

Architected an AI-driven automation framework using Playwright and a custom MCP server that autonomously converts product tickets into executable UI and API tests.

Developed a standalone AI agent (decoupled from the IDE) capable of:

• Generating structured test cases and production-ready POM/spec files
• Launching browser sessions to execute tests end-to-end
• Performing self-healing element resolution until successful execution
• Reusing existing framework utilities via an internal memory layer to prevent code duplication

Designed custom tooling to orchestrate the workflow and deployed the agent as a containerised service using Docker, enabling execution within CI/CD pipelines and reducing test creation time to ~3 minutes from 1 hour 35 minutes

Built an internal RAG chatbot !

using embeddings + a vector database to search and answer questions over company documents and Confluence pages.

• Created UI & API Automation Framework using Playwright/JavaScript
• Created Visual Regression Testing using AI (OpenAI API) -saved the company roughly £60k -they were planning to use Apitools
• Created Performance Testing Framework - using K6
• Created Helper functions that use API requests for UI tests
• Implemented best practice for Framework Design Pattern involving inheritance and abstraction for multiple languages
• Configured CI/CD Pipeline to run as a kron job and runs on code merges
• Test Report Notifications from pipeline will show results on MS Teams and Testrail and Azure DevOps
• Mentored Junior QA engineers to automate and upskill

• Implemented shift left testing with a TDD approach for our QA process
• Created CSharp api framework using Playwright- ran various tests from previous cypress framework
• Wrote YAML files to run the automation as part of the CI/CD pipeline - configured to store artifacts on S3 Bucket as well as the implementation of GithubActions Triggers to run Automation Suites because of separate repositories
• Created UI Automation Framework using Cypress and completed all of the happy paths which boasts 250 test cases
• Created API Automation Framework using Cypress and implemented the Integration Tests which now has over 700 test cases
• Created Performance Testing Framework using K6- which can now simulate 1000 virtual users for Load/Stress/Spike testing
• Successful migration to Kubernetes from EC2
• Mentored Senior and Junior QA's to learn and utilise our automation framework
• Carry out Line Management Duties for the QA's and help them successfully achieve their goal in terms of learning and being an excellent team members for their teams.
• Developed and executed comprehensive test strategies for different squads
• Managed test planning, resource allocation, and test case design for multiple projects
• Have regular weekly meetings with C-Level Executives to discuss the performance and quality of our product

• Created Kubernetes Clusters for Test Automation using TestKube which was then pulling the test cases as part of the CI/CD pipeline

• Created brand new UI Automation Framework from scratch using JS and Cypress

• API automation with CSharp And RestSharp
  

• Mentoring and teaching junior members on how to use Cypress

• Built a Bitbucket Pipeline to run for every pull request and connecting both the separate repositories

• Successfully completed my tasks for UI and API automation – both have been brought to current sprint level

• Set up framework and introduced Percy for visual testing

• Maintained, debugged and optimized automation framework.

• API automation with Postman and Newman

• Set up a Test Management suite for manual testing and coach the best practices

• Designed, developed, and maintained automated test suites for a Shopify-based eCommerce platform using tools such as Cypress
• Used Shopify’s Admin API, Storefront API, and Liquid templates to validate backend functionality and data consistency in automated tests.
• Collaborated with developers, designers, and product managers to define clear acceptance criteria and ensure seamless integration of new features.
• Implemented automated regression, smoke, and end-to-end tests across critical user journeys including product browsing, checkout, payment processing, and order fulfillment.
• Built and maintained CI/CD pipelines (e.g., GitHub Actions, Jenkins, or GitLab CI) to automatically run test suites on every deployment.
• Conducted cross-browser and cross-device testing to ensure a consistent customer experience across desktop and mobile environments.

• Working remotely, and managing QA teams across UK, Croatia, and USA.
• Provide QA Sign-Off and approval on iOS code releases. Draft and maintain internal QA documentation on QA tools and best practices.
• Mentor and onboard junior-level QA Engineers.
• Design test strategy for existing and new features
• Using breakpoints for destructive testing, network throttling
• SSL debugging – allowing decryption of encrypted data to review/troubleshoot the transmitted content.
• Build automation Strategy in adherence to a well balanced test pyramid and drive improvements to app performance, usability and hardware/software compatibility
• Drive incident prioritization and work cross-functionally with engineering/design
• Write and maintain automated test procedures using Swift and XCUI
• Work hand in hand with Design, PM team and engineering teams to ensure quality feature delivery
• Participate in release processes for the entire company
• Work closely with Customer Operations, Localization, and other internal teams to triage incidents and drive resolution when necessary
• Provide technical support to internal team members
• Host and manage internal company test sessions
• Test video streaming for live shows during air/and pre-rehearsals
• Verified bug fixes for optimal system performance.

Festicket(NFC/RFID apps for transactions) The company pioneers in events and festivals.

• I test their backend as well as frontend, for compatibility, bugs, UI and their mobile app. Which is Cashless, Cahsless is an easyway for the clients of festicket to enjoy the event; Cashless POS(point of sales) solution lets client load credit onto branded RFID wristbands to pay for drinks, food and merchandise with a quick tap.
• Overseen two events that have both achieved more than 80k transactions in one week, across 3 different NFC devices.
• Test using the MIFARE tags with NFC devices for transactions, balance, customer refunds.
• Maintain, create test coverage, for each new version of the apps released for the different types of NFC devices and different events.
• Test their event management portal to see backend compatibility.
• Exploratory testing, for potential problems via transaction without/poor internet connection.

• Set up BDD for SDLC.
• Set up test tools and tests to be used for the company and best practices.
• API Testing using Postman.
• Exploratory testing( there are so many tests that can be mentioned here, but it would be very lengthy).
• Troubleshoot/Triaging bugs/reproducing bugs reported by clients across various devices( iOS and Android).
• Creating (regression and brand new feature suits).
• Liaising on release builds with developers to ensure timely fixes for prioritised bugs, with weekly bug meeting with the Senior iOS and Senior Android devs

• In charge of the QA process, from writing test cases from Confluence to testing before releasing to the live market.
• Backend compatibility testing.
• UI testing across various devices on both platforms.
• Managing release builds, and test planning time on dev-builds bug fixes.
• Reproducing bugs on test environment from production environment with creating same scenarios as live User.
• Testing on both platforms for IOS and Android.
• Leading the liaising with QA/Dev's across Europe.

• Working as Test Analyst delivering cutting edge mobile apps for both clients and the company itself, supporting delivery of projects throughout the software development life cycle.
• Writing and executing test scripts/ test cases.
• Exploratory testing of applications.
• Management of client expectations as well as the product owner.
• Raising defects using JIRA