MLOps Engineer

We are looking for a talented and driven MLOps Engineer to join our AI/ML team. In this role, you will design and build scalable ML infrastructure, MLOps pipelines, and CI/CD workflows for deploying and managing NLP, generative AI, and computer vision models.
Responsibilities:
ML Infrastructure and Platform: Design, build, and maintain scalable MLOps pipelines and infrastructure for training, deploying, and managing NLP, GenAI, and computer vision models.
CI/CD for Machine Learning: Implement automated continuous integration and continuous deployment (CI/CD) pipelines for ML models, ensuring smooth transitions from development to production environments.
Model Serving and Deployment: Lead the deployment and lifecycle management of Large Language Models (LLMs), Vision-Language Models (VLMs), and other ML models using scalable serving engines (e. g., vLLM, Triton Inference Server, Ray Serve).
Monitoring and Observability: Set up comprehensive monitoring frameworks to track model performance, data drift, resource utilization, and system health in real-time production environments.
Pipeline Orchestration: Develop and manage end-to-end ML workflows encompassing data ingestion, preprocessing, training, evaluation, and inference using tools like Kubeflow, MLflow, or Apache Airflow.
Cloud and Resource Optimization: Optimize cloud infrastructure (AWS/GCP/Azure) for cost, latency, and throughput. Ensure efficient hardware resource allocation across GPUs and distributed computing clusters.
Cross-Functional Collaboration: Collaborate closely with data scientists, NLP/CV engineers, and software engineers to define hardware constraints, capacity planning, and deployment strategies.
Requirements:
Qualification: BTech in CS/AI/Data Science or related discipline.
Possess 2-5 years of professional experience in MLOps, machine learning engineering, or closely related technical domains.
Demonstrated expertise in the deployment and lifecycle management of Large Language Models (LLMs), VLMs, and diverse ML architectures within production-grade environments.
High proficiency with MLOps and serving frameworks, including MLflow, Kubeflow, vLLM, Triton Inference Server, and TensorRT for high-performance inference.
Practical experience implementing vector databases (such as Pinecone, FAISS, or Weaviate) alongside robust CI/CD tooling like GitHub Actions, Jenkins, or Terraform.
Advanced coding skills in Python, C++, and Bash, with a strong background in managing containerized workloads using Docker and Kubernetes across major cloud platforms (AWS, GCP, or Azure).
Comprehensive understanding of ML pipelines and observability, including model versioning, registries, and monitoring frameworks like Prometheus and Grafana for tracking system health and data drift.
Excellent communication abilities with a collaborative mindset and a focus on solving complex technical challenges.

Experience
2-5 yrs