Senior Agentic Ai Engineer Sapiens Bengaluru

Vibehackers; Bengaluru; On-site; INR 3,000,000 - 6,000,000; Full time

Job summary

Sapiens in Bengaluru seeks a Senior Agentic AI Engineer to design, build, and productionize agentic AI systems that automate complex, document-heavy insurance workflows.

You will own agent architecture and orchestration, develop end-to-end RAG pipelines, ensure reliability and safety, and integrate with enterprise tools to deliver auditable, production-grade automation. Collaborate across teams to deliver robust Python code and maintainable solutions.

Qualifications

• Experience building production agentic AI systems for enterprise workflows.
• Hands-on with LangGraph or equivalent orchestration frameworks.
• Proficiency in Python with production-grade coding, testing, and documentation.
• Experience with end-to-end RAG systems: indexing, retrieval, grounding, evaluation.

Responsibilities

• Design and implement agentic systems capable of multi-step reasoning and workflow execution across lifecycles.
• Build stateful workflows using LangGraph or equivalent orchestration frameworks.
• Engineer for long-horizon reliability: recovery from errors and planning under uncertainty.
• Develop end-to-end RAG pipelines: ingestion, chunking, embeddings, retrieval, grounding strategies.
• Implement observability, tracing for prompts, tool calls, and production behavior.
• Apply guardrails and safety controls to reduce hallucinations in live settings.
• Evaluate agents with tests, regression analyses, and automated checks.
• Build integrations with enterprise tools, APIs, databases, and model providers.

Skills

Agent architecture

Orchestration

RAG pipelines

Observability

Reliability engineering

Testing

CI/CD

Documentation

Safety controls

Tools

LangGraph

Claude (Anthropic API)

Azure AI Foundry

Job description

Builds and orchestrates agentic AI systems (LangGraph, MCP) and uses AI assistants like Claude - directly focused on agent-driven, AI-assisted engineering.

About the Role

Senior Agentic AI Engineer at Sapiens in Bengaluru to design, build, and productionize agentic AI systems that automate complex, document-heavy insurance implementation workflows. The role focuses on agent architecture and orchestration, RAG pipelines, reliability and safety, and integrations with enterprise systems to deliver auditable, production-grade automation.

Role

Senior Agentic AI Engineer responsible for designing and implementing production-grade agentic systems to automate complex, document-intensive insurance implementation processes. The role covers agent architecture and orchestration, retrieval and grounding, reliability and safety, evaluation, and integrations with enterprise tools and model providers.

Key Responsibilities

• Design and implement agentic systems capable of multi-step reasoning, planning, tool use, and workflow execution across implementation lifecycles.
• Build stateful workflows (branching, retries, self-correction, human-in-the-loop checkpoints) using LangGraph or equivalent orchestration frameworks.
• Engineer for long-horizon reliability: multi-step task completion, recovery from compounding errors, planning under uncertainty.
• Develop end-to-end RAG pipelines: ingestion, chunking, embeddings, vector/hybrid retrieval, reranking, contextual compression, and grounding strategies.
• Implement conversational state and persistent memory, retrieval-aware context assembly, and token-efficient context selection.
• Apply MCP-style tool and context interfaces for agent access to enterprise knowledge repositories and structured configuration data.
• Implement observability and tracing for prompts, tool calls, retrieval quality, agent traces, failures, drift, latency, and production behavior.
• Apply guardrails, safety controls, and failure-handling to reduce hallucinations in agents used in live client settings.
• Evaluate agents at trajectory and task level with sandboxed tests, regression analysis, automated checks, and human review.
• Build integrations with internal/external tools, APIs, enterprise systems, databases, and model providers.
• Deliver production-quality Python code with strong testing, CI/CD, logging, versioning, and documentation practices.
• Translate ambiguous implementation processes into robust system logic and reusable AI patterns.

Requirements

• Demonstrated experience building and shipping production agentic AI systems.
• Strong, hands-on experience with LangGraph or equivalent agentic orchestration frameworks, including custom orchestration.
• Deep proficiency in Python with production-grade code, testing, and documentation practices.
• Experience designing and optimizing end-to-end RAG systems: indexing, retrieval, reranking, grounding, and evaluation.
• Daily working proficiency with Claude (Anthropic API) and Claude Code.
• Experience building and deploying agents on Azure AI Foundry or equivalent enterprise cloud AI platforms.
• Practical understanding of LLM behavior, hallucination risks, reasoning constraints, and evaluation methods.
• Experience evaluating and debugging agent behavior at trajectory and task level.
• Hands-on experience with MCP-based interoperability patterns and tool-calling agent design.
• Modern software engineering practices for LLM systems: testing, CI/CD, observability, tracing, and debugging.

Preferred Qualifications

• Experience with multi-agent orchestration and agent collaboration patterns.
• Familiarity with vector databases such as Pinecone, Weaviate, Azure AI Search, OpenSearch.
• Experience processing complex, unstructured document types (contracts, RFPs, configuration files, regulatory documents).
• Exposure to model adaptation techniques like LoRA or QLoRA.
• Prior experience in insurance, financial services, or enterprise SaaS implementation environments.
• Habit of staying current with AI research, benchmarks, and emerging engineering patterns.

Skills

Agent architecture Orchestration System design Reliability engineering Context engineering Retrieval-Augmented Generation (RAG) Observability and tracing Testing CI/CD Debugging Integration Documentation Safety and risk mitigation Human-in-the-loop design Evaluation and metrics