AI Systems Engineer specializing in production-grade LLM pipelines, hybrid RAG architectures, and multi-agent orchestration. Dual background in Computer Science Engineering and Applied Mathematics — rigorous theory paired with hands-on deployment.
Currently building scalable AI systems at bld.ai — from graph-enhanced retrieval and LangGraph agents to full-stack production apps.
| Role | Company | Period |
|---|---|---|
| AI Engineer & Full Stack Developer | bld.ai · Remote | May 2023 – Present |
| Data Engineering Intern | Tecnoquímicas S.A. · Cali | Jan 2022 – Jul 2022 |
Highlights at bld.ai
- Architected hybrid RAG pipelines — dense vector search (Qdrant) + Neo4j knowledge graphs for multi-hop reasoning and reduced hallucinations
- Designed multi-agent orchestration with LangGraph: task decomposition, dynamic tool routing, graph-based state management
- Fine-tuned transformer models for domain-specific classification; evaluated with BLEU, BERTScore, and perplexity
- Built production backends (FastAPI, PostgreSQL, MongoDB) and frontends (React, TypeScript)
- Integrated MCP servers and external APIs into autonomous, zero-interruption agent workflows
|
2024 – Present Production multi-agent platform with LangGraph — graph-based state, structured tool routing, and fully autonomous execution modes. Agents integrate with external APIs, MCP servers, and notification systems.
|
2024 – 2025 Conversational AI combining Qdrant vector search with Neo4j graphs for multi-hop reasoning over structured and unstructured data. BM25 + dense hybrid retrieval for a Spanish tax-domain chatbot.
|
|
2023 – 2024 Fine-tuned transformers for multi-class classification in industrial settings. Experiment tracking with MLflow; evaluation via BLEU, BERTScore, and perplexity.
|
2021 – 2022 · repo Social-network-enhanced recommender using NetInf cascade inference, graph features (PageRank, centrality), and Collective Matrix Factorization. Peer-reviewed publication ↓
|
Uribe, S., Ovalle, C., Finke, J. (2023). Recommender Systems Based on Matrix Factorization and the Properties of Inferred Social Networks. Discrete Mathematics, Algorithms and Applications.
| MSc in Artificial Intelligence · Universidad Internacional de La Rioja | 2023 – 2024 · GPA 8.4/10 |
| BSc Computer Science Engineering & Applied Mathematics · PUJ Cali | 2017 – 2023 · CS 4.14/5 · Math 4.55/5 |
| AWS Cloud Practitioner Essentials | Jan 2024 |
| Microsoft Certified: Azure Fundamentals | Jan 2024 |
| Area | Tools |
|---|---|
| AI / LLM | LangChain · LangGraph · Hybrid RAG (BM25 + dense) · Fine-Tuning · Tool-Calling Agents · MCP |
| Data & Infra | Apache Airflow · ETL Pipelines · MLflow · Qdrant · Cassandra |
| Languages | Python · TypeScript · JavaScript · SQL · R · C++ |
Building AI systems that are reliable, explainable, and maintainable at scale.