I specialize in building end-to-end AI systems, from serverless data pipelines to production-grade ML models. My work focuses on automating complex workflows using Google Cloud Platform (GCP), Vertex AI, and Large Language Models.
Governed Multi-Agent Smart City Orchestrator Live Demo | View Repository
An enterprise-grade decision engine implementing a rigorous "Advocate vs. Critic" architecture with Constitutional AI patterns, ensuring AI-generated infrastructure proposals are safe, compliant, and hallucination-free.
- Stack: Google ADK, Gemini Pro, Vertex AI Search, FastAPI, Next.js, Firestore
Live Platform | View Repository
A fully production-grade AI financial analysis platform on Google Cloud. Generates actionable trading signals for the Russell 1000 by automating the ingestion of financial data, analyzing it using custom ML models, and surfacing structured Call/Put setups.
- Stack: GCP, Vertex AI, Python, BigQuery, Cloud Run
AI-powered financial intelligence server built on the Model Context Protocol. Integrates GCP BigQuery and LLMs to automate high-gamma options strategies and market structure analysis.
- Tech: Model Context Protocol, BigQuery, Python, GCP, FastMCP
A secure file vault enforcing strict user isolation and irreversible PII redaction using event-driven architecture and Cloud DLP.
- Tech: Cloud Run, Cloud DLP, Vertex AI, FastAPI, React
AI-powered alcohol label compliance checker automating verification via Gemini 2.0 Flash and Cloud Vision multimodal analysis.
- Tech: Gemini 2.0, Cloud Vision, FastAPI, Cloud Run
arXiv-powered paper recommender using Gemini embeddings and Vertex AI Vector Search with A/B testing support.
- Tech: Gemini Embeddings, Vertex AI Vector Search, Cloud Run, FastAPI
Multi-document scientific paper analyzer providing RAG-powered Q&A with citations using Vertex AI and Gemini.
- Tech: RAG, Vertex AI, Gemini, FastAPI, Firestore
Cloud-native content generation pipeline using a Map-Reduce architecture to automate large-scale literary analysis.
- Tech: Vertex AI, Gemini 3 Flash, Python, Map-Reduce
A comprehensive guide for fine-tuning YOLOv9 on custom datasets, from preparation to inference.
- Tech: PyTorch, YOLO, Computer Vision, Deep Learning
| Domain | Tools |
|---|---|
| AI & Orchestration | Google ADK, Vertex AI, Gemini, OpenAI, TensorFlow, PyTorch, YOLO |
| Cloud & DevOps | Google Cloud Platform (GCP), Cloud Run, Cloud Functions, BigQuery, Docker |
| Backend | Python, FastAPI, Node.js, Firebase |
| Frontend | React, Next.js, TypeScript, Tailwind CSS, Astro |
- Portfolio: evanparra.ai
- LinkedIn: linkedin.com/in/evanparra
- Twitter: @ParraEvan
- Email: admin@evanparra.ai