Not a requirement for an Architect but needs someone who can work hands on and should have 10-15 years of exp in IT. Immediate Available to start this fully remote role.

We're looking for a Senior Agentic AI Engineer to design, build, and produce LLM-powered agents that run reliably at enterprise scale. This is a hands-on role for someone who has moved beyond prototypes and demos — you've shipped agentic systems that real business users depend on, and you understand what it takes to make them observable, governed, and trustworthy.

You'll sit at the intersection of applied AI, data engineering, and platform design. Beyond building individual agents, you'll help shape the reusable foundations — agent skills, a semantic layer, and data-quality guardrails — that let the rest of the organization build agentic applications faster and more safely.

What You'll Do

- Design and build multi-agent and single-agent systems for enterprise use cases, from problem framing through production deployment and iteration.
- Architect agent orchestration flows (planning, tool use, memory, human-in-the-loop, error recovery) using LangGraph, LangChain, and DeepAgents.
- Instrument, trace, and evaluate agents in production using Langfuse and/or LangSmith — building the observability and evaluation loops that catch regressions and quantify quality.
- Author reusable agent skills and tools that encapsulate domain logic and can be composed across applications, with clear contracts, versioning, and documentation.
- Design and maintain the semantic layer that agents query — defining metrics, entities, relationships,



and business definitions so agents reason over trustworthy, consistent data.
- Build and harden the data pipelines feeding agentic systems, embedding data-quality checks (validation, freshness, lineage, anomaly detection) as first-class concerns.
- Select, evaluate, and integrate foundation models; design RAG, tool-use, and fine-tuning strategies appropriate to each use case, balancing cost, latency, and accuracy.
- Partner with data science to bring experimentation rigor to agent behavior — offline and online evaluation, A/B testing, and metric-driven iteration.
- Collaborate with security, platform, and compliance teams to meet enterprise requirements around access control, data governance, PII handling, and auditability.

Core Requirements

- Proven experience building and shipping agentic applications in production, not just prototypes — ideally within an enterprise or regulated environment.
- Robust hands-on expertise with LangGraph and LangChain, including complex orchestration patterns (state machines, conditional routing, memory, tool calling).
- Experience with DeepAgents or comparable frameworks for deep / long-horizon agentic workflows.




- Practical experience with LLM observability and evaluation using Langfuse and/or LangSmith — tracing, prompt/version management, and building evaluation datasets and scorers.
- Solid working knowledge of AI / foundation models: prompting, RAG, tool use, context management, and the tradeoffs between model families and deployment options.
- Strong software engineering practices: version control, testing, CI/CD, code review, and writing maintainable production code (Python expected).

Data & Platform Skills

We're specifically looking for depth across the foundations that make agentic systems reliable:

Data Engineering — designing and operating batch and streaming pipelines (ETL/ELT), working with modern data warehouses/lakehouses, and orchestration tooling.

Data Quality — implementing validation, testing, lineage, monitoring, and anomaly detection so downstream agents consume trustworthy data.

Semantic Layer Design — modeling metrics, entities, and business definitions; experience with semantic/metrics layers, ontologies, or knowledge graphs that agents can query consistently.

Agent Skill Authoring — designing reusable, well-documented agent skills and tools with clear interfaces, guardrails, and versioning for reuse across teams.

Background in MLOps / LLMOps — deployment, monitoring, and lifecycle management of models and agents.

Experience in a regulated or highly governed industry (finance, healthcare, life sciences, etc.).

Contributions to open-source agentic or data tooling.

📌 Senior Agentic AI Engineer (Chennai)
🏢 Judge India Solutions
📍 Chennai