Senior AI/ML & Agentic Infrastructure Engineer - Alltegrio (Multi-agent Enterprise AI Platform)
Duration: 05/2024-08/2026
Summary:
- Architected a multi-agent enterprise AI platform on Kubernetes using Python and VertexAI, enabling autonomous task orchestration and improving workflow efficiency
- The platform supports generative AI and machine learning workloads across multiple cloud providers with high availability
Responsibilities:
- Designed and implemented containerized retrieval and inference services with semantic search and scalable APIs.
- Established reusable CI/CD pipelines for model validation, image builds, security checks, and staged releases.
- Operationalized AI workloads across GCP, AWS, and Azure with resilient multi-cloud deployment patterns.
- Built post-deployment model monitoring for quality, latency, throughput, and infrastructure health.
- Optimized distributed training and inference workflows in PyTorch and TensorFlow.
- Owned architecture and production troubleshooting for Kubeflow pipelines integrating data preparation, model training, evaluation, approval, and deployment.
- Collaborated with cross-functional teams on design reviews, documentation, and production standards.
Technologies: Python, Kubernetes, VertexAI, Docker, GitHub Actions, Kubeflow, PyTorch, TensorFlow, GCP, AWS, Azure
Senior MLOps & Data Platform Engineer - Mesh-Al (Cloud-Native Data Mesh and ML Infrastructure Modernization)
Duration: 11/2021 - 04/2024
Summary: Led modernization of a cloud-native data mesh spanning GCP, AWS, and Azure, providing governed data products and ML infrastructure to accelerate model delivery and improve operational efficiency.
Responsibilities:
- Built distributed processing pipelines with Apache Beam and Dataflow for feature preparation and batch inference.
- Implemented standardized Kubeflow, VertexAI, and SageMaker workflows for training, evaluation, registration, and production promotion.
- Created automated CI/CD pipelines with GitHub Actions and Docker for model deployment.
- Developed model monitoring workflows for prediction quality, drift, latency, and throughput.
- Optimized autoscaling and resource allocation for containerized inference services on Kubernetes.
- Collaborated with data owners, data scientists, software engineers, and governance stakeholders to define platform standards and resolve production issues.
Technologies: Kubeflow, SageMaker, Apache Beam, Dataflow, VertexAI, GitHub Actions, Docker, Kubernetes, GCP, AWS, Azure
Machine Learning Engineer Cloud & MLOps - ML6 (Production Machine Learning Services and Pipelines)
Duration: 07/2019 – 10/2021
Summary: Developed production machine learning services and end-to-end training and deployment pipelines on GCP and VertexAI, improving prediction latency and availability for cloud-based ML applications.
Responsibilities:
- Developed ML services in Python using TensorFlow, PyTorch, and scikit-learn.
- Built training and deployment pipelines on GCP and VertexAI.
- Containerized inference applications with Docker and deployed on Kubernetes.
- Implemented CI/CD validation for code, data schemas, model artifacts, and deployment configurations.
- Created scalable feature-processing workflows with Apache Beam and Dataflow.
- Partnered with data scientists and software engineers to troubleshoot models and pipelines and establish monitoring practices.
Technologies: Python, TensorFlow, PyTorch, scikit-learn, GCP, VertexAI, Docker, Kubernetes, Apache Beam, Dataflow
Data Scientist & Predictive Analytics Engineer - Tremend (Predictive Analytics and Real-Time Fraud Detection Pipelines)
Duration: 10/2014 - 07/2017
Summary: Built predictive analytics pipelines and a real-time fraud detection workflow processing millions of daily transactions, delivering low-latency predictions and improving model evaluation and throughput.
Responsibilities:
- Developed predictive analytics pipelines in Python and scikit-learn.
- Implemented real-time fraud detection workflow with low-latency predictions.
- Built data preparation, training, and evaluation components with TensorFlow.
- Optimized feature extraction and scoring services for production workloads.
- Collaborated with software engineers, QA, and business stakeholders to define model acceptance criteria and document production support procedures.
Technologies: Python, scikit-learn, TensorFlow
Upstaffer since September, 2026
