5 days ago
Responsibilities
- Design end-to-end multi-agent architectures that automate data engineering tasks such as pipeline orchestration, data quality remediation, schema drift detection, anomaly resolution, and metadata management.
- Define agent patterns for planning and reasoning loops, tool use, memory and state management, multi-agent orchestration, and human-in-the-loop checkpoints.
- Select and evaluate agent orchestration frameworks for enterprise-scale deployment.
- Establish standards for LLM integration, prompt and context management, retrieval-augmented generation, and function or tool calling.
- Partner with data engineering teams to identify automation opportunities across ingestion, transformation, orchestration, and observability.
- Architect agents that analyze metadata, logs, and lineage to detect, diagnose, and appropriately remediate pipeline failures.
- Integrate automation solutions with data warehouses, lakehouses, ETL/ELT tools, orchestration engines, and catalogs.
- Define guardrails, evaluation frameworks, reliability and safety monitoring, cost and latency controls, human oversight, escalation, and rollback patterns.
- Lead technical design reviews, proofs of concept, pilot-to-production transitions, and mentoring for engineering teams.
Requirements
- 15+ years of experience in software, data, or AI architecture roles, including enterprise-scale system design.
- At least 2 years of hands-on experience designing or implementing agentic AI or LLM-based systems, with production or near-production exposure preferred.
- Foundational experience with data engineering, including pipeline design, ETL/ELT, orchestration tools, and data quality frameworks.
- Strong understanding of LLM architectures, prompt engineering, retrieval-augmented generation, embeddings, vector stores, and function or tool calling.
- Experience with agent orchestration frameworks such as LangGraph, AutoGen, CrewAI, Semantic Kernel, or equivalent custom orchestration layers.
- Familiarity with Airflow, dbt, Spark, Kafka, Snowflake, Databricks, BigQuery, and Redshift.
- Cloud architecture experience with AWS, Azure, or GCP, including AI and machine learning services.
- Understanding of API design, microservices, event-driven architectures, and MLOps/LLMOps practices.
- Preferred qualifications include DataOps automation experience, exposure to Collibra, Alation, or Unity Catalog, cloud or AI/ML certifications, and experience with agent evaluation, guardrail, testing, or red-teaming frameworks.
Tech Stack
Amazon RedshiftApache AirflowApache KafkaApache SparkAWSAzureDatabricksdbtGoogle BigQueryGoogle Cloud PlatformSnowflake
Categories
AI ApplicationsData Engineering
About Cognizant
Cognizant (Nasdaq: CTSH) is an AI Builder and technology services provider, bridging the gap between AI investment and enterprise value. We build full-stack AI solutions powered by deep industry, process and engineering expertise — embedding an organization's unique context into technology systems that amplify human potential and drive tangible outcomes. From strategy to deployment, we help global enterprises move from AI ambition to AI impact and stay ahead in a fast-changing world. See how at cognizant.ai | Follow us @cognizant