Ramona ramonacraft

👩‍💻 About Me

Engineering & delivery leader for web, iOS, Android, and CTV — I turn AI into practical QA and release systems teams can actually run. Automate the rinse-and-repeat. Keep humans on judgment.

class DeliveryLeader:
    def __init__(self):
        self.name = "Ramona Bonitatis"
        self.role = "Engineering & Delivery Leadership"
        self.focus = "AI Transformation · QA Engineering"
        self.location = "North Los Angeles 🌴"
        self.platforms = ["Web", "iOS", "Android", "CTV"]

    def skills(self):
        return {
            "day_to_day": ["Cursor", "Claude", "Playwright", "MCP", "Jira"],
            "delivery": ["Release judgment", "P0-P3 focus", "CI/CD gates"],
            "building": ["TestMCP", "ForgeQA", "Release Gate Lab", "War Room"],
            "learning": ["Databricks", "Prompt evals", "LangGraph"],
        }

    def how_i_work(self):
        return [
            "Automate the rinse-and-repeat",
            "Keep humans on edge cases + judgment",
            "Lead with visuals so teams align fast",
            "Ship practical AI systems, not hype",
        ]

🎯 Quick Snapshot

📍	Based in	North Los Angeles
💼	Focus	Engineering & Delivery Leadership · AI Transformation · QA Engineering
📈	Impact I've led	635 Playwright tests across 7 suites · web smoke ~90% faster (hours → minutes)
🔥	Building with	Cursor, Claude, Playwright, MCP, Docker, RAG / vector search, Azure Pipelines, React/TypeScript
🌱	Learning next	Databricks, prompt eval frameworks, LangGraph, deeper TypeScript, stronger CI/CD
💡	Philosophy	Automate the rinse-and-repeat. Keep humans on judgment.
🤝	Works with	Engineering · Product · Project · QA Engineering · BI (Analytics)

"15+ years across web, iOS, Android, and CTV — shipping fast with judgment, learning in public, and building systems teams can actually run."

🚀 What I build in public

These projects show how I think about Agile, shipping fast, release risk, live windows, and AI-assisted QA Engineering — project management, technical integration, QA practices, and a builder mindset.

	Project	What it is
👩‍💻	Live Event War Room	Live-event ops board + living runbook (real signals + Healthy → Incident)
🚦	Release Gate Lab	Lean Playwright gate + go/no-go dashboard · live demo
🧪	TestMCP	PR changes → risk-ranked P0–P3 test plan (MCP + dashboard)
⚒️	ForgeQA	Local-first AI → reviewable Playwright from a plain-English goal

🔀 How the work connects

flowchart LR
  A["🧪 TestMCP<br/>PR risk"] --> B["⚒️ ForgeQA<br/>generate tests"]
  B --> C["🚦 Release Gate<br/>lean Playwright"]
  C --> D["👩‍💻 War Room<br/>live window"]

  style A fill:#F8FAFC,stroke:#111827,color:#111827
  style B fill:#FEE2E2,stroke:#DC2626,color:#111827
  style C fill:#E5E7EB,stroke:#374151,color:#111827
  style D fill:#FCE7F3,stroke:#BE185D,color:#111827

PR risk → test generation → lean gate → live window view

🛠️ Tech Stack & Tools

💻 Languages & platforms

✅ Working with now

🌱 Learning next

Inspired by Andrew Ng and DeepLearning.AI — practical AI systems, evals, and building with judgment, not hype.

💡 How I Work


🔁 Automate the repeat	Keep humans on edge cases, validating AI agent work, and release judgment.
🎯 Trusted suites	Prefer small P0–P3 suites over giant suites that quietly get deprioritized.
📊 Fresh priorities	Quarterly reviews with Product and Business Intelligence mapped to KPIs.
🎨 Visual first	Boards, flows, and dashboards over long walls of text.
📚 Always learning	New skills and tools, then put them to work with the team. Monitor token usage, optimize for AI agent loops, and prompt less.

✨ Currently

🚦 Shipping the public delivery loop end to end — Release Gate live demo sits in the middle of PR risk → tests → lean gate → live window
👩‍💻 Building portfolio labs that mirror real release and live-ops decisions (War Room, TestMCP, ForgeQA)
🧠 Learning hands-on with a local second brain (search by meaning + Claude) so the AI tools I recommend are ones I've actually run

Navigation Menu