Selected work

Sole data engineer for a platform delivering evidence-based longevity research.

• Designed and implemented the full PubMed ingestion pipeline in Python.
• Built a RAG architecture: semantic search with vector embeddings + LLM reranking for a clinical-answer chatbot.
• Managed all GCP infrastructure with Terraform: Cloud Storage, BigQuery, Cloud Run and Vertex AI.
• Owned pipeline reliability, data-quality monitoring and iterative improvement of the retrieval system.

Predicting biological age from scRNA-seq data, in both regression (age in months) and classification (young vs. adult).

• Processed large-scale scRNA-seq datasets with scanpy, such as Tabula Muris Senis.
• Trained and evaluated regression (Elastic Net, Random Forest, XGBoost) and classification (Logistic Regression, Random Forest, SVM) models.
• Feature selection via highly variable genes and correlation with age.
• Ran all compute on the IBMB HPC cluster with SLURM in reproducible workflows.

Data support for the business: reliable reporting and transformation processes for KPIs and analytical models.

• Automated SQL reporting workflows and Python preprocessing pipelines.
• Implemented validation and data-quality controls across the processes.
• Designed data transformations feeding analytical models and business KPIs.