I bring a research-oriented yet practical approach — combining experimentation, optimization, and systems thinking — to develop scalable, interpretable, and high-performance AI solutions aligned with real-world applications.
I hold strong fundamentals (ML, evaluation, calibration) + functional thinking (risk scoring, text quality scoring, feature classification, preventing/forecasting) to deliver robust solutions that mirror today’s industry needs.
Experience Level
Language
Work Experience
Education
Qualifications
Industry Experience
This project exhibits a Coordinator Agent (Block A) orchestrates Risk / Fraud / Payoff
sub-agents y sync with a Decision Platform (RedHatPAM) that encapsulates rules, decision tables,
and governance, with a migration path to Kogito for event-driven microservices and ephemeral RHPAM where appropriate.
Why it matters: Design aligns with modern data products (low latency, backpressure
tolerance, replayability), while BRMS provides explainability, policy,
and audit required by stakeholders.
An interpretable multi-criteria NLP model:
Hybrid rules + transformers with stable cross-criterion performance and audit trail. Includes also optimization strategies for transformer-based models, exploring adaptive learning rates, gradient clipping, and mixed precision training to enhance performance in natural language processing applications.
This is also a focused POC/MVP/ prototype system for implementing multi-task learning architectures using DeBERTa models, covering fine-tuning strategies, loss balancing, and evaluation metrics for complex NLP scenarios.
Hire a AI Engineer
We have the best ai engineer experts on Twine. Hire a ai engineer in Santiago today.