I'm a Computer Science student and aspiring MLOps Engineer specializing in production-grade machine learning systems. My work focuses on deploying models reliably, implementing full-stack observability, and optimizing infrastructure for ML workloads. I approach infrastructure with a software engineering mindset: strictly typed, version-controlled, and thoroughly tested.
Production-grade Retrieval-Augmented Generation system with Langfuse v3 tracing, FlashRank reranking, and Prometheus monitoring. Achieved 94% recall and near-zero hallucination rate.
Stack: Python, FastAPI, LangChain, Qdrant, Docker, Kubernetes
View Code Read Docs
End-to-end MLOps pipeline with Airflow orchestration, MLflow tracking, and automated monitoring. Features data drift detection, model evaluation, and Grafana dashboards.
Stack: Airflow, MLflow, Docker Compose, Prometheus, Grafana, MinIO
View Code Read Docs
Production microservice serving Transformer models on constrained hardware (1GB RAM). Features Nginx reverse proxy, Redis caching, and automated CI/CD with 99.5% uptime.
Stack: FastAPI, Docker, AWS EC2, Nginx, Redis, GitHub Actions
View Code Read Docs
Exploring LLM fine-tuning, vector database optimization, and Kubernetes operators for ML workloads. Check my GitHub for the latest updates and experiments.
Focus: LLM Fine-tuning, K8s Operators, Infrastructure as Code
View GitHubI'm actively seeking Junior MLOps Engineer positions where I can build production ML infrastructure, implement observability solutions, and design scalable RAG systems. Open to remote opportunities and relocation.