Building production AI systems — LLM & agentic workflows, RAG, computer vision, and edge AI. Published computer-vision researcher (Springer).
- 🔭 Currently: Head of AI at Tripolar Industries — Physical / Edge AI for industrial facilities (digital twins, MPC-based optimization, agentic assistants).
- 🛠️ I work with: LLM inference & serving (vLLM, quantization), agentic workflows (LangGraph), RAG & GraphRAG, vector search (Pinecone, ChromaDB), and production data pipelines.
- 📄 Published research: Enhancing License Plate Recognition in Videos Through Character-Wise Temporal Combination — Springer, Cham (2023).
- 🏆 Recipient of Cuba's highest national innovation award for applied AI research.
- 💬 Ask me about: LLM serving & inference optimization, agent design with LangGraph, RAG/GraphRAG, computer vision.
- 📫 Reach me: cquiala12work@gmail.com
- Semantic Product Search Backend — ETL pipeline for product indexing with embeddings (Sentence Transformers, all-MiniLM-L6-v2) and Pinecone vector search for fast semantic similarity.
- Local Semantic File Search (in progress) — privacy-first CLI that finds documents by meaning with exact citations; runs fully local and exposes an MCP server for agent integration.
- Computer Vision — License Plate Recognition — CV solution backing my Springer publication.
AI / ML: LLMs · Generative AI · RAG · GraphRAG · Agentic Workflows (LangChain, LangGraph) · Fine-tuning · MPC · Deep Learning (PyTorch, TensorFlow) · Computer Vision (OpenCV) · scikit-learn
LLM Inference & Serving: vLLM · Hugging Face · Quantization · Ollama · OpenAI / Gemini APIs
Vector Search: Sentence Transformers · Pinecone · ChromaDB · semantic search
Data & MLOps: Python · SQL · FastAPI · PySpark · Pandas · Apache Airflow · Docker · Kubernetes · Git/CI-CD · BigQuery
Cloud: AWS · GCP · Azure (AI-900)
