👋 Surya Kulshreshtha - About Me
SDET · Test Automation Architect — Python • Playwright • TypeScript • CI/CD
🇮🇳
| Info | Details |
|---|---|
| 🚀 Current Endeavour | Python + Automation using Playwright · AI engineering from first principles · DevOps from setup to full send |
| 🔬 Latest Build |
|
| 🤝 Collaboration | Looking to collaborate on projects |
| 📧 Contact |  ,  ,  |
| 📂 Repositories | 10 Forkable* repos, mapped in the Repository Index below |
⭐ If a
Forkable*repo saves you an afternoon, a star helps someone else find it.
Everything I publish is a Forkable* repository: clone it, run it, keep it or throw it away. This page is the map — what each repo is, what stack it uses, and the order to read them in.
| Repo | What it is | Stack |
|---|---|---|
| ForkablePythonProjectShowcase | End-to-end test automation framework — UI + API, Page Object Model, network mocking, Docker, GitHub Actions CI | Python Playwright Pytest Docker |
| ForkablePlaywrightSelfHealer | Self-healing Playwright tests written from plain-English specs; repairs broken locators offline, heuristics first with optional local LLM | Python Playwright |
| Repo | What it is | Stack |
|---|---|---|
| ForkablePythonLearningShowcase | Test automation fundamentals, built up step by step | Python |
| Forkable04EyePythonPlaywrightLearning | E2E testing workflows — sample suites wired to GitHub Actions CI | Python Playwright |
| Forkable04EyeLearningSandbox | Scratchpad for testing practitioners — experiments, patterns, spikes | Python |
| Repo | What it is | Stack |
|---|---|---|
| ForkableFullStackQASuite | Dual-language test suite for an employee CRM — REST API, role-based visibility, MSSQL ground truth, Chart.js data checks; two teams, one shared test plan | TypeScript Python Playwright MSSQL |
| ForkableTypescriptPostgreReactNodejsLearning | Full-stack playground — an app worth testing, front to back | TypeScript React Node.js PostgreSQL |
| ForkableInterviewToolkit | Interview prep for QA/SDET roles, with runnable proof rather than bullet points | TypeScript Python |
| Repo | What it is | Stack |
|---|---|---|
| Forkable-DevOps-Complete-StarterPack-To-FullSend | Nine phases from workstation setup to a capstone, built around one small service (FullSend): browser tests, CI/CD, containers, Kubernetes, Terraform, SLOs and alerts, supply-chain security, and a read-only CI triage agent. Commands for macOS and Windows | Python FastAPI Playwright Docker Kubernetes Terraform GitHub Actions |
| Repo | What it is | Stack |
|---|---|---|
| Forkable-AI-Engineering-Starter-Pack-To-Full-Send | Nine levels, 28 drills: autograd, tokenizers, attention, BM25 retrieval, agents and evals, each judged by tests. The capstone is a RAG engine that answers with citations or refuses | Python stdlib only RAG Agents |
Forkable04EyeLearningSandbox → poke at things, no stakes
ForkablePythonLearningShowcase → learn the primitives
Forkable04EyePythonPlaywrightLearning → put a suite in CI
ForkablePythonProjectShowcase → the full framework
ForkableFullStackQASuite → two teams, two languages, one test plan
ForkablePlaywrightSelfHealer → make it survive the app changing
ForkableInterviewToolkit → talk about all of it convincingly
Forkable-DevOps-Complete-StarterPack-To-FullSend → ship it, run it, keep it observable
Forkable-AI-Engineering-Starter-Pack-To-Full-Send → build the AI you'll be testing
Languages · Python · TypeScript · JavaScript · SQL
Testing · Playwright · Pytest · Page Object Model · API & contract testing · network mocking
App/Backend · Node.js · Express · FastAPI · React · Next.js · Chart.js
Data · PostgreSQL · MongoDB · MSSQL
AI · RAG · BM25 · autograd · attention · agents · LLM evals
Ops · Docker · GitHub Actions · Git · GitLab · Postman · Kubernetes · Helm · Argo CD · Terraform · Prometheus · Grafana
- Framework architecture — layering, fixtures, test data strategy, keeping suites maintainable past 500 tests
- Flake elimination — deterministic waits, isolation, retries that don't hide bugs
- Self-healing automation — locators that survive UI churn without a human in the loop
- CI/CD for tests — parallelisation, containerised runs, reporting that people actually read
- API + UI in one suite — setting state through the API, asserting through the UI
- Multi-team test governance — one shared test plan and fixture set, independently implemented in TypeScript and Python so two teams catch different blind spots
- DevOps from setup to full send — one small service taken through tests, CI/CD, containers, Kubernetes, Terraform, SLOs and supply-chain security, with Playwright browser tests in the loop (DevOps Starter Pack)
- AI engineering from scratch — autograd, attention, retrieval and agents in plain Python, every piece judged by tests (Full Send)
These predate the automation work by about a decade. They stay because they were useful once, and the "how do I start" one still is.
| File | Why it's still here |
|---|---|
| How do I learn programming & subsequently software development? | Still the answer I'd give someone starting out |
| Mac OS X Terminal Basic Commands | The terminal basics I still reach for |
| Rock Paper Scissors in Python | The first thing I wrote here. It stays. |
Test automation framework collaboration · QA architecture reviews · mentoring people moving into SDET work · anything involving making a flaky suite trustworthy.
📬 · 🌐 suryakulshreshtha.in