Skip to content
View rayperatta's full-sized avatar
🥸
🥸

Block or report rayperatta

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
rayperatta/README.md

Ray Peratta

Operations professional in Lisbon, building practical AI-assisted automation projects with Python, APIs and workflow tools. I’m moving toward AI implementation, operations automation and technical integration roles.

My background is in client operations at Accenture and Conectys. The repositories below are independent portfolio projects: a place to demonstrate implementation decisions, test failure cases and document what still needs work.

Selected projects

GitHub-to-Zendesk API integration with signed webhooks, a durable outbox, bounded retries and an authenticated recovery API. Includes a local HTTP failure lab for duplicate events, rate limits and lost receipts, plus automated tests. Start here for integration reliability and technical support workflows. Live Zendesk validation remains a separate step.

Text intake, structured extraction, validation, SQLAlchemy persistence and rule-based routing with FastAPI and Prefect. Includes a mock provider and sample documents for local development. Start here for the document workflow and its current limitations.

A C#/.NET reporting project with CSV ingestion, anomaly checks, Excel/PDF generation, scheduling and a dashboard. Email delivery has a demo mode without SMTP credentials. Sample data supports a reporting walkthrough.

A Python/FastAPI prototype for PDF question answering, combining keyword and vector retrieval with reranking. Includes optional tracing, an evaluation harness and documented next steps.

How I work

I use AI-assisted development to build and iterate. My focus is making workflows understandable: clear inputs and outputs, validation, repeatable demos, meaningful tests and explicit limitations.

Technologies used across these projects: Python, SQL, REST APIs, FastAPI, n8n, PostgreSQL, SQLAlchemy, Docker and C#/.NET.

Let’s connect

Open to AI implementation, operations automation and technical integration opportunities, and narrowly scoped automation projects.

Lisbon, Portugal · Authorized to work in Portugal
Spanish — native · English — C1 · Portuguese — B2

LinkedIn · Email

Pinned Loading

  1. ai-document-intake ai-document-intake Public

    Python/FastAPI document workflow prototype: structured extraction, validation, persistence and routing with a mock provider and tests.

    Python

  2. dailyops-bot dailyops-bot Public

    .NET reporting prototype: CSV ingestion, anomaly checks, Excel/PDF reports, scheduling and a sample-run dashboard.

    C#

  3. rag-document-qa rag-document-qa Public

    PDF question-answering prototype with hybrid retrieval, reranking, optional tracing and a RAGAS evaluation harness.

    Python

  4. ops-insight ops-insight Public

    C#/.NET operations reporting engine: raw CSV export to SLA & backlog report, a code-first alternative to Power Automate + Power BI

    Python