Agentic AI and Data Engineer

Agentic AI and Data Engineer Location: Hyderabad, India (hybrid; subject to confirmation) Experience: 5+ years in software or data engineering, including 2+ years of hands-on AI agent development Role Type: Hands-on Engineering / Individual Contributor Role Overview We are seeking an Agentic AI and Data Engineer to build, integrate, and operate AI agents and the data foundations that support them. The ideal candidate combines strong software engineering and data engineering skills with practical experience developing agentic workflows, retrieval and grounding systems, tool integrations, and evaluation mechanisms. The role involves building the agentic capabilities that power enterprise product experiences and ensuring those capabilities are grounded in reliable, relevant, and accessible data. The candidate should understand the trade-offs between different agent architectures and have experience moving beyond prototypes toward reliable, observable, production-oriented systems. Key Responsibilities Agentic AI Engineering • Design, implement, and maintain AI agents that interact with tools, APIs, enterprise data, and other application services. • Build agentic workflows using appropriate patterns, including single-agent, multi-agent, planner/executor, routing, and human-in-the-loop designs. • Implement tool/function calling, orchestration, context and memory management, retrieval, and task execution. • Select suitable agent architectures based on use-case complexity, reliability requirements, latency, cost, and operational constraints. • Integrate agents with enterprise applications, APIs, knowledge sources, and structured or unstructured data. • Implement error handling, retries, fallbacks, access controls, and other safeguards for dependable agent execution. Data Engineering and Grounding • Build and maintain data pipelines that ingest, transform, validate, and prepare data for AI agents. • Develop retrieval and grounding sources that improve the relevance, traceability, and factual consistency of AI-generated responses. • Integrate agents with enterprise APIs, product data, knowledge graphs, and other relevant data sources. • Work with data models, metadata, entity relationships, and semantic structures to support accurate information retrieval. • Implement data quality checks, schema validation, freshness monitoring, and appropriate error handling. • Support the preparation and indexing of data for retrieval-augmented generation (RAG) and related grounding approaches. Evaluation, Reliability, and Observability • Develop and maintain evaluations for agent behavior, task completion, response quality, groundedness, and tool usage. • Measure and improve reliability, latency, cost, and output quality through iterative testing and optimization. • Instrument agent workflows to capture relevant execution traces, failures, latency, and usage metrics. • Investigate failures such as incorrect tool calls, retrieval gaps, hallucinations, context loss, and workflow loops. • Collaborate with SDET and other engineering team members to establish regression tests for agent workflows. • Apply appropriate safety checks and human review mechanisms to sensitive or high-impact workflows. Engineering and Collaboration • Write maintainable, modular, testable code and participate in code reviews. • Collaborate with the Technical Lead, frontend developers, SDET, and subject-matter experts to deliver end-to-end capabilities. • Translate product and domain requirements into agent workflows, data contracts, and implementation tasks. • Document architecture, integration contracts, evaluation results, and operational considerations. • Keep up with evolving agentic AI patterns and apply them pragmatically to real product requirements. Required Skills and Experience • 5+ years of experience in software engineering, data engineering, backend engineering, or a closely related discipline. • At least 2 years of practical, hands-on experience building AI agents or agentic AI workflows. • Strong programming skills in Python or another suitable backend language. • Practical experience with tool/function calling, agent orchestration, context management, retrieval, and evaluation. • Understanding of single-agent and multi-agent architectures, planner/executor workflows, routing, and human-in-the-loop patterns. • Experience building and maintaining data pipelines and integrating multiple data sources. • Understanding of RAG, embeddings, vector retrieval, data indexing, and grounding techniques relevant to the solution. • Experience integrating APIs and handling data formats, schema changes, authentication, and failures. • Strong software engineering fundamentals, including testing, debugging, version control, and code quality. • Ability to measure and improve agent quality rather than relying only on subjective demonstrations. Preferred Qualifications • Experience integrating knowledge graphs, graph databases, ontologies, semantic layers, or oData services. • Experience with agent evaluation frameworks, tracing, observability, and production monitoring. • Familiarity with cloud-based AI services, data platforms, and deployment pipelines. • Experience with LLM application frameworks or orchestration libraries. • Experience building or fine-tuning ML models; this is a differentiator, not a mandatory requirement. • Experience deploying and operating AI agents used by real users in production environments. Expected Outcomes • Functional agentic workflows aligned with defined product use cases. • Reliable data pipelines and grounding sources supporting those workflows. • Appropriate evaluation and regression mechanisms for agent quality. • Measurable improvement in reliability, response quality, latency, and cost. • Effective integration with enterprise APIs, product data, and knowledge sources. • Well-documented, testable, and maintainable engineering deliverables. Ideal Candidate Profile A hands-on engineer who has shipped AI agents beyond a proof of concept and understands the complete lifecycle—from data preparation and retrieval through orchestration, evaluation, deployment, and monitoring. The candidate should be able to explain why a particular agentic pattern was chosen, how the data grounds the agent, and how the solution's quality and reliability are measured.