Mayur Patil

Mayur Patil · Engineer, Pune

I build the platforms that let teams ship with confidence.

Engineering platforms, developer tooling and AI agents for software that has to be right the first time. Ten years in regulated finance, currently at Barclays.

See the work GitHub LinkedIn
85%faster release validation
100+specs migrated to Playwright
1,400tests on a self-built grid
~150reports on one platform

Systems, not scripts.

Every project here replaced manual effort with something other engineers now build on.

AI-driven development lifecycle

Agents that turn source repos into test suites and audit evidence.

Claude Code agents connected to Jira, Confluence and GitLab over MCP read the application code, generate test cases and runtime data, and produce rule-level traceability that holds up in audit.

Repo + Confluence→ Claude Code agent→ Tests + data→ Evidence
Developer platform

A Playwright framework teams adopt.

Page objects, data-grid helpers, environment configs, DB fixtures, CI templates. Now the team standard.

import { test } from './fixtures'; test('report ready', async ({ grid, db }) => { await grid.filter({ status: 'Final' }); await db.expectRows('report', 1); }); // fixtures: env, DB, grid helpers
Agentic ops

Natural-language control of services, jobs and pods.

An MCP server over an internal Spring Boot API so one request can chain restarts, Autosys jobs and OpenShift checks across environments.

prepare the UAT environment and check pod health
Done. services restarted · 3/3 pods healthy · data refresh job queued
Release engineering

Regression from six hours to forty-five minutes.

A parallel-execution platform across SIT, UAT and pre-prod that made same-day release-candidate validation routine.

Before6 h
After45 m
Performance tooling

Load tests generated from the real system.

Swagger specs and recorded user flows become Karate-Gatling simulations, so performance tests track the API rather than a hand-written guess.

Swaggeror HAR capture
→
GeneratorKarate features
→
Gatlingsimulation
Data validation at scale

Reusable reconciliation utilities for millions of records.

A library of recon and validation utilities that compare datasets across AWS Athena, Oracle and MSSQL, recompute expected results independently, and check generated Word, PDF and Excel outputs. Millions of rows reconciled in hours, with row-level diffs as evidence.

Athena12.4M rows
⇄
SQL12.4M rows
0 mismatches · 1h 52m

From one test grid to the platform a bank relies on.

Each role shipped something the team still uses after I moved on to the next problem.

2021 – present

AVP, Product Test Engineer

Barclays · Pune · investment-banking regulatory reporting
  • Cut release regression from 6 h to 45 min (85%) with a parallel-execution platform; same-day release-candidate validation across SIT, UAT and pre-prod.
  • Migrated a 100+ spec Protractor suite to Playwright + TypeScript in 12 weeks; the framework is the team standard today.
  • Introduced an AI-driven development lifecycle: Claude Code agents over MCP generate tests and evidence from source repos and sync results to Xray.
  • Reconcile millions of records in minutes across AWS Athena, Oracle and MSSQL with a reusable validation library.
  • Mentor engineers on Playwright, Karate and AI-assisted testing; review framework code across the programme.
2019 – 2021

Automation Engineer

Western Union · Pune · global payments compliance
  • Replaced manual regression with a BDD framework covering REST APIs, desktop and web, run in-sprint with results posted to Slack.
  • Built CI/CD pipelines that deploy Java and .NET microservices to cloud behind automated quality gates.
  • Caught a critical performance defect before go-live with LoadRunner; co-founded WU QAfé, the internal engineering-quality community.
2016 – 2019

Systems Engineer

Tata Consultancy Services · Pune
  • Cut web regression from two weeks to three days by architecting a Selenium Grid that ran 1,400 tests across nodes with automated Jira reporting.
  • Built BDD mobile automation for Android and iOS on Perfecto, and hybrid web frameworks on Selenium, Cucumber and TestNG.
B.E. Electronics & Telecommunication
Savitribai Phule Pune University · 2012 – 2016 · First Class with Distinction

Toolkit.

Full-stack across the test pyramid, the delivery pipeline and the AI layer on top.

Languages

TypeScriptJavaScriptJavaPythonC#SQL

Frameworks

Playwrightplaywright-bddKarateCucumberSpecFlowWebdriverIOSelenium GridAppiumRest AssuredTestNGpytest

Performance

GatlingKarate-GatlingLoadRunnerJMeter

AI

Claude CodeGitHub CopilotGitLab DuoMCP serversAgents & skillsLLM test generation

Platform

Spring BootOracleMSSQLGitLab CIJenkinsDockerOpenShiftAWS

Certified

ISTQBSAFe 4 PractitionerRPA Developer

Open source.

Frameworks and experiments on GitHub.

quantum-test-frameworkHybrid Java framework for web, mobile, API and desktop
bdd-protractor-tsUI automation with parallel execution and Docker
webdriverio-cucumber-tsWebdriverIO + Cucumber in TypeScript
mayurpatild.github.ioThis site. One HTML file, no build step

Let's build something.

Open to senior engineering and platform roles, and to conversations about AI-assisted engineering.

Open mail Download resume
Copied mayurpatild@gmail.com