I build AI and data systems for financial services: agent workflows, retrieval design, model operations, and governed data platforms built for real enterprise constraints.
Production AI, banking automation, and operating models that make advanced systems usable after launch.
Building agent workflows that help analytics teams retrieve, reason over, and review enterprise knowledge with traceable outputs.
Created the MLOps policies, procedures, manuals, and dashboards used to monitor production models, then extended the practice into an LLM Operational Framework.
AI workflows should leave reviewable paths, not black-box confidence.
Models and agents need monitoring, procedures, owners, and escalation paths.
Enterprise AI has to work with governance, messy data, and teams who need trust.
The technical range behind the work: AI design, data platforms, operating discipline, and user-facing systems.
Agent workflows, MCP, retrieval design, and enterprise context engineering.
Python, SQL, Spark, Databricks, Airflow, model evaluation, and dashboards.
MLOps, LLMOps, drift monitoring, procedures, governance, and reliability.
Banking support workflows, secure webchat, Messenger flows, NLP, and validation.
Before production AI, I worked through language modeling, sentiment analysis, molecular simulation, and evidence-led experimentation.
Filipino tweet sentiment models and molecular dynamics simulation with bioinformatics collaborators.