I build ML platforms that help data scientists ship models to production faster. Machine learning models don’t fail because of algorithms they fail because the infrastructure around them is fragile, slow, or impossible to scale. My work focuses on solving that problem. I’m a Machine Learning Platform Engineer with 3+ years of experience designing production-grade ML systems — from feature stores and training pipelines to large-scale inference and model lifecycle management. I enjoy building systems that make machine learning reliable, reproducible, and easy to operate at scale. 🔹 MLOPS/ML platform I design end-to-end ML infrastructure that enables data scientists to move from experimentation to production seamlessly. This includes building: • Feature stores and reusable feature pipelines • Automated training and retraining systems • Model promotion workflows with GitOps • Scalable batch and inference pipelines • Experiment tracking and model lifecycle management 🔹 Scaling Machine Learning Systems I specialize in building ML systems that can handle real production workloads. Examples of systems I’ve built: • Batch inference pipelines processing 200M+ records per day • Drift-triggered retraining systems reducing compute by ~75% • CI/CD pipelines that reduced ML rollout time from weeks to days • Data-driven orchestration that cut pipeline runtime by 40%+ 🔹 Cloud Native ML Infrastructure Most of my work lives at the intersection of machine learning, infrastructure, and developer tooling. I regularly work with: • Kubernetes-based ML infrastructure (EKS/GKE) • Infrastructure-as-Code (Terraform / Pulumi) • GitOps workflows and CI/CD pipelines • Databricks, Snowflake, and distributed data systems • Observability stacks for ML workloads 🔹 What I Care About The ML ecosystem still lacks great developer experience for machine learning systems. I’m particularly interested in building: • ML platforms • Feature stores • LLM infrastructure • developer tooling for ML teams • scalable reasoning systems 💡 Core Technologies & Keywords GCP • AWS Python • PyTorch • HuggingFace • Scikit-Learn • Pandas • PySpark • Polars Kubernetes • Docker • Helm • ArgoCD Terraform • Pulumi • Infrastructure-as-Code MLflow • ZenML • Dagster • Airflow • Metaflow • Feast Databricks • Snowflake • Delta Lake • DuckDB AWS (EKS, S3, EC2, Lambda, SQS) • GCP (GKE, GCS, Cloud run) • Azure GitHub Actions • Bitbucket Pipelines • CI/CD • GitOps Prometheus • Grafana • Loki • Observability Feature Stores • ML Pipelines • Model Serving • Large-Scale Inference