Design and develop multi-agent AI orchestration pipelines and AI agent workflows.

- Build Human-in-the-Loop (HITL) workflows with approval, review, and escalation mechanisms.
- Develop and maintain LLM integration layers / Model Abstraction Layers.
- Perform advanced Prompt Engineering, including system prompts, few-shot prompting and structured outputs.
- Build RAG (Retrieval Augmented Generation) pipelines using embeddings and vector databases.
- Integrate Azure OpenAI / Vertex AI models through managed APIs.
- Develop deterministic tools for REST APIs, database queries, file parsing, validation, and enterprise data retrieval.
- Implement robust error handling, retry mechanisms, token management, and LLM failure mitigation.
- Build audit-ready logging covering agent actions, evidence retrieval, decisions, and outputs.
- Containerize applications using Docker and deploy on Kubernetes (AKS/GKE).

MUST-HAVE SKILLS

Python Expert level

LLM / Generative AI application development

Agentic AI / Multi-Agent Systems

Prompt Engineering

RAG / Embeddings / Vector Databases

Azure OpenAI and/or Vertex AI

REST API / OpenAPI / API Integration

PostgreSQL / Cloud SQL

Docker & Kubernetes (AKS/GKE)

Git / CI/CD

AGENTIC AI EXPERIENCE

Candidates should have hands-on experience with concepts such as:





ReAct | Tool Calling | Multi-Agent Orchestration | Agent State Management | HITL | Structured Outputs | Prompt Injection Mitigation | LLM Evaluation

Experience with LangChain, LlamaIndex, AutoGen, or similar agentic frameworks is an advantage.

SECURITY & ENTERPRISE EXPERIENCE

Good understanding of

- Encryption at rest/in transit
- Secrets management & Key Vault / Secret Manager
- Least-privilege access
- Audit logging & traceability
- Data residency
- Secure API integrations

• Cloud-native security

Experience in BFSI / Financial Services / regulated environments is highly preferred.

GOOD TO HAVE

- MLOps / LLMOps
- Model versioning & regression testing
- Blue-Green deployments
- Azure Confidential Computing / GCP Confidential VMs
- TypeScript / Node.js

• Experience deploying AI solutions in regulated environments

IDEAL PROFILE We are looking for a hands-on AI/LLM Engineer, not a traditional Data Scientist or research-only profile.

The ideal candidate should have experience taking LLM/Agentic AI solutions from development to production, with robust focus on reliability, security, auditability, and enterprise integration.

Interested candidates can share their updated CV at:

[[email protected]]

📌 AI Developer | Agentic AI & LLM Engineering (Greater Noida)
🏢 NLB Services
📍 Greater Noida