Work Experience
Databricks - Senior MLOps EngineerNew York
09/2023 - Present
ML Platform Team - Enterprise MLOps
Led MLOps initiatives:
- Architected end-to-end ML platforms serving 2000+ data scientists across 100+ enterprise clients
- Developed automated CI/CD pipelines for ML model deployment reducing deployment time by 90%
- Implemented real-time model monitoring system processing 10M+ predictions daily
- Built distributed feature store handling 100TB+ of feature data
- Created automated model governance and compliance frameworks
- Key technologies: MLflow, Kubeflow, Airflow, Kubernetes, Prometheus
Achievements:
- Reduced model deployment cycle from weeks to hours
- Improved model monitoring coverage to 100%
- Automated 95% of MLOps workflows
- Implemented cost optimization saving $3M annually
- Achieved SOC 2 and ISO 27001 compliance
Weights & Biases - MLOps Platform EngineerSan Francisco
08/2020 - 08/2023
ML Infrastructure - Experiment Tracking & Model Registry
- Developed scalable experiment tracking system
- Built automated model versioning and registry
- Implemented reproducible ML pipelines
- Created model performance monitoring
- Developed artifact management system
Platform Development
- Designed MLOps best practices
- Implemented security controls
- Built collaboration features
- Created custom visualizations
- Developed API integrations
Achievements:
- Scaled platform to handle 1M+ experiments
- Reduced experiment setup time by 75%
- Improved platform reliability to 99.99%
Technical Skills
MLOps Tools
- MLflow, Kubeflow
- Weights & Biases
- DVC, ClearML
- Seldon Core
- BentoML
Infrastructure
- Kubernetes, Docker
- Terraform, Ansible
- CI/CD (Jenkins, GitLab)
- ArgoCD, Flux
- Service Mesh
Monitoring & Observability
- Prometheus, Grafana
- ELK Stack
- Datadog, NewRelic
- Jaeger, OpenTelemetry
- Custom Metrics
Development
- Python, Go
- SQL, NoSQL
- REST APIs, gRPC
- Git, GitHub Actions
- Cloud Platforms
ML Engineering
- Model Deployment
- Feature Stores
- A/B Testing
- Model Versioning
- Pipeline Orchestration
Education
Columbia University - Master of ScienceComputer Science
09/2006 - 07/2010
- GPA: 3.95/4.0
- Thesis: 'Scalable MLOps Platforms for Enterprise AI'
- Published papers on ML systems and infrastructure
- Research Focus: ML Systems and DevOps
- Relevant Coursework: Distributed Systems, Cloud Computing, ML Engineering, DevOps Practices
- Led MLOps Community (450+ members)
- Created automated ML pipeline framework adopted by research groups
Open Source & Projects
MLOps Automation Framework
Enterprise MLOps platform
- 14k+ GitHub stars
- Used by 300+ companies
- 2.5M+ downloads
- Featured in InfoQ and The New Stack
- Supports GitOps workflows
- Implements ML-specific CI/CD
Model Monitoring Suite
Production model monitoring system
- 9k+ GitHub stars
- Real-time monitoring capabilities
- Featured in MLOps Community
- Advanced alerting system
- Drift detection algorithms
Feature Store Framework
Distributed feature management
- 8k+ GitHub stars
- Used in production environments
- Real-time feature serving
- Advanced caching strategies