Embedded Infrastructure Engineer, Chanakya

About this job

Sarvam AI is building India's full-stack sovereign AI platform, focusing on research, models, infrastructure, and applications to make AI work for India. As an Embedded Infrastructure Engineer for Chanakya at Sarvam AI in Delhi, you will design, build, and maintain the data infrastructure that supports AI system deployments at client locations. Working closely with Embedded Data Scientists and Strategic Deployment Engineers, your daily routine involves creating and managing databases, object stores, ingestion pipelines, and processing layers for terabyte-scale datasets. You will make key choices regarding storage architecture, indexing, pipeline orchestration, and system performance, especially in air-gapped, on-premise, or sensitive environments where standard cloud services might not be available. Additionally, you will monitor platform health, ensure reliability, and handle capacity planning for assigned accounts.

Responsibilities

• Design and operate data storage architectures managing terabyte-scale datasets across various modalities.
• Build and maintain ingestion pipelines for daily data influx with monitoring and error handling.
• Implement indexing, partitioning, and query optimisation strategies for low-latency data retrieval.
• Collaborate with Embedded Data Scientists to translate schemas and semantics into physical data models.
• Deploy and manage database systems and vector stores in security-constrained or air-gapped environments.
• Build observability into data platforms to monitor pipeline health, storage, and query performance.
• Own capacity planning and scaling decisions for data infrastructure across assigned client deployments.

Requirements

• 4–8 years of experience in data infrastructure, data engineering, platform engineering, or site reliability engineering.
• Direct experience managing multi-terabyte data stores handling 10TB+ of persistent data.
• Deep working knowledge of at least two of: PostgreSQL, MongoDB, Elasticsearch, or ClickHouse.
• Experience building production data ingestion pipelines using Apache Kafka, Apache Spark, Airflow, Flink, dbt, or equivalent frameworks.
• Strong proficiency in Python and/or Go for writing infrastructure tooling and automation.
• Solid understanding of storage systems like object storage and columnar formats.
• Familiarity with containerisation and orchestration tools like Docker and Kubernetes in production.
• Experience with infrastructure-as-code and deployment automation tools like Terraform.

Good to have

• Experience with vector databases or embedding stores like Milvus, Weaviate, Qdrant, or pgvector.
• Experience deploying and operating infrastructure in air-gapped, on-premise, or hybrid environments.

Who should apply

This role suits candidates with 4 to 8 years of solid background in data engineering, platform engineering, site reliability engineering, or data infrastructure who have managed large-scale multi-terabyte systems. If you have hands-on expertise in Python or Go, data pipelines, and database tuning, and you are motivated to solve challenging infrastructure problems in sensitive or air-gapped environments for AI deployments at Sarvam AI, you should apply.

Frequently asked questions