Job Summary

We are looking for an AI/ML Engineer – Agentic AI with strong experience in Python, LLMs, LangGraph/LangChain, RAG, multi-agent systems, and prompt engineering.

The role focuses on building and deploying autonomous AI agents, enterprise copilots, and intelligent workflows using Azure/AWS, Docker, Kubernetes, and MLOps/AgentOps, with an emphasis on scalability, security, observability, and responsible AI.

Key Technical Skills

1. Core Programming & Systems

- Expert-level Python for ML, orchestration, and agent logic
- Solid understanding of async programming, concurrency, and task scheduling

2. Agentic AI

- Autonomous AI agents for multi-step reasoning, planning, task orchestration, and decision-making
- ReAct, Plan & Execute, Reflexive agents
- Hierarchical and multi-agent systems
- Tool-augmented and function-calling agents
- Stateful/stateless agents and memory management

3. Large Language Models

- Hands-on experience with OpenAI, Azure OpenAI, Anthropic, or open-source LLMs
- Prompt engineering, reasoning, planning, critique, and tool-use techniques
- Few-shot and zero-shot prompting
- Model selection based on latency, cost, and context length
- Fine-tuning/LoRA knowledge is a plus

4. Agent Frameworks & Tooling

- LangGraph / LangChain
- Semantic Kernel
- AutoGen, CrewAI, or similar frameworks
- Custom agent orchestration
- Tool abstraction, execution safety, retries, and timeouts

5. RAG, Memory & Knowledge Augmentation

- Short-term and long-term agent memory
- RAG implementation
- Vector databases such as FAISS, Pinecone,



Azure AI Search, etc.
- Embeddings, chunking, and context management
- Knowledge graph/hybrid memory is a plus

6. Planning, Reasoning & Control

- Task planning and re-planning
- Constraint-based execution
- Feedback and self-correction loops
- Guardrails, action validation, failure detection, and recovery

7. MLOps / AgentOps

- Production deployment of AI agents
- Agent observability, tracing, logging, and monitoring
- Model and prompt versioning
- CI/CD for AI/agent systems
- Docker, Kubernetes, and Azure/AWS deployments

8. Evaluation & Testing

- Agent evaluation frameworks
- Task success, cost, latency, reliability, and safety evaluation
- Hallucination detection
- A/B testing of prompts, tools, and agent strategies

9. Security & Responsible AI

- Secure tool execution and privilege control
- Prompt injection and jailbreak mitigation
- Data privacy and memory isolation
- Human-in-the-loop escalation
- Explainability and responsible AI practices

10. Data & Integration

- Enterprise system, API, CRM, ERP, database, and SaaS integrations
- SQL / NoSQL
- Event-driven systems and message queues are a plus

11. Cloud & Platform

- Strong experience in at least one cloud platform: Azure, AWS, or GCP
- Managed AI services
- IAM, secrets management, and security
- LLM cost optimization

12. Nice to Have

- Multi-agent collaboration
- Enterprise copilots and autonomous workflows
- Human-AI collaboration patterns
- Reinforcement learning for agent optimization

📌 AI/ML Engineer (Chennai)
🏢 Prophecy Technologies
📍 Chennai