Senior MLOps Engineer

Spectral Consultants is hiring for a Senior MLOps Engineer (GenAI / LLM) with one of the leading US Product based organization.

Location : Noida.

Work Mode : Remote.

Experience : 6-10 years.

The candidate is responsible to build and scale production-ready ML & GenAI (LLM) systems on AWS, enabling seamless deployment, monitoring, and optimization of AI models.

Key Responsibilities :

• Design, build, and scale ML/AI platforms on AWS, leveraging services such as SageMaker, Bedrock, S3, Lambda, ECS/EKS.

• Develop and deploy LLM-based applications, including RAG pipelines, embeddings, and vector search solutions.

• Build and manage end-to-end ML pipelines (data ingestion - training - deployment - monitoring).

• Implement CI/CD pipelines for ML and LLM systems, ensuring automated testing, validation, and deployment.

• Deploy and manage real-time and batch inference systems with high scalability and reliability.

• Monitor model performance, drift, and bias using tools like CloudWatch, Prometheus, and Grafana.

• Work with vector databases such as OpenSearch, Pinecone, or FAISS.

• Collaborate with Data Scientists, ML Engineers, and cross-functional teams to productionize ML models.

• Ensure platform security, governance, and cost optimization aligned with best practices.

Candidate Requirements :

• 6+ years of experience in ML Engineering, MLOps, or related domains.

• Strong hands-on experience with Generative AI, LLMs, and RAG pipelines.

• Expertise in AWS ecosystem, including SageMaker, Bedrock, S3, Lambda, ECS/EKS.

• Experience with model deployment, monitoring, and ML lifecycle management.

• Strong programming skills in Python and experience with APIs (FastAPI/Flask).

• Hands-on experience with Docker, Kubernetes, and Infrastructure-as-Code (Terraform/CloudFormation).

• Experience working with vector databases (OpenSearch, Pinecone, FAISS).

• Solid understanding of CI/CD pipelines for ML systems.

• Strong problem-solving, communication, and collaboration skills.

Salary : Hidden