Data Engineer & AI Researcher — building intelligent data systems, end-to-end ML pipelines, and generative AI solutions that turn complex data into strategic decisions.
Data Engineer & AI Engineer specialized in RAG systems, autonomous agents, and data pipelines. I build end-to-end solutions from legacy data ingestion to conversational AI interfaces, formally evaluating each technique with precision and recall metrics.
Currently pursuing a Master's in Data Science at Universidad Nacional de Ingeniería (UNI), with a background in Mechatronics Engineering, Environmental Engineering, and Project Management. I lead a private AI research laboratory focused on search architectures, evaluation frameworks, and graph-based anomaly detection.
Lima, Peru · 882 followers on LinkedIn · 500+ connections
Demonstrates that data quality improvements (ground truth correction, embedding deduplication) yielded a 27% improvement in Precision@5 — with zero code changes and zero cost — outperforming neural reranking approaches. Based on 28 formal evaluations with 52 reference questions.
Design and prototype of an autonomous two-wheeled mobile robot using inverted pendulum control systems, with stress simulations and material validation.
Graph Neural Networks with temporal pattern analysis for detecting fraud rings in instant payment networks. Using 6.3M synthetic transactions modeled after regional patterns.
Type two sentences and see how AI measures their meaning similarity using embedding vectors. Powered by open-source models via HuggingFace Inference API.
Dual embedding (384d + 768d) with vector cosine similarity, full-text search, and Reciprocal Rank Fusion. Intelligent query routing based on complexity.
1,345 entities and 1,716 relationships extracted from domain documents. Enables relationship-aware queries beyond keyword matching.
52 ground truth questions across 10+ domains, 62 evaluation runs. Metrics: Precision@5, MRR, Hit Rate, Recall with stratified difficulty levels.
End-to-end pipeline: image capture → preprocessing → ML classification → PDF report → email delivery. Anemia detection from nail images.
Every system follows secure development principles: credential rotation, environment isolation (dev/staging/prod), encrypted secrets management, and automated QA checks before deployment.
Rate limiting per IP, CAPTCHA verification, request fingerprinting, anomaly detection on access patterns, and automatic circuit breakers for suspicious behavior.
Distributed architecture for processing natural language queries (NLQ) to democratize BI in enterprise environments. Asynchronous WebSocket orchestration with adaptive timeouts; semantic engine using open-source LLMs for NLQ-SQL translation enriched with ontological schemas (50+ business terms) and hierarchical fallbacks. Conversational interface with persistent context and distributed tracking.
MSc Data Science · Universidad Nacional de Ingeniería (UNI)
2024 — 2026 · Big Data, Python, SQL, R, Data Analysis
MSc Project Management · Escuela de Postgrado UTP
2021 — 2023
BSc Mechatronics Engineering · Universidad San Ignacio de Loyola (USIL)
2020 — 2023
BSc Environmental Engineering · Universidad San Ignacio de Loyola (USIL)
2015 — 2019
Diploma in Agile Project Management · CENTRUM PUCP
2022 · Agile Product Management, Coaching, Agile Teams
Diploma in Biomedical Equipment Management · TECH SENATI
2021 — 2022 · Clinical & Hospital Engineering
Whether you're a recruiter, researcher, or fellow engineer — I'd love to hear from you. Reach out through any of these channels:
Lima, Peru · Open to collaborations, research partnerships, and new opportunities.