Description

You will design and orchestrate agentic workflows to advance AI capabilities through intelligent, multi-model decision-making.

This role is on-site.

Responsibilities

• Evaluate and integrate orchestration frameworks to optimize model routing logic.
• Configure AI agents to select the best model for each query using GPT-4, Llama 2, and Claude.
• Develop fallback and escalation mechanisms for tasks lacking confidence or accuracy.
• Fine-tune multi-step agentic workflows incorporating APIs, RAG, and reflection prompts.
• Monitor performance metrics and AI decision policies, refining workflows based on LLM evaluation metrics.

Required Skills

• 5–8+ years of experience in AI, Machine Learning, NLP, or AI Systems Engineering.
• 2–3+ years building AI agents, designing model routing architectures, or working with multi-model AI orchestration.
• Expertise in agentic frameworks such as LangChain, LlamaIndex, CrewAI, or OpenAI function calling.
• Deep knowledge of LLM architectures (GPT-4, Claude, Llama 2, Mistral) and multi-model workflow orchestration.
• Strong understanding of Retrieval-Augmented Generation (RAG) and vector database integration.
• Experience designing self-assessment and reflection prompts to improve agent reasoning accuracy.
• Proficiency in Python, OpenAI SDK, Hugging Face, and cloud-based AI deployments.
• Strong grasp of LLM safety mechanisms, fallback strategies, and ethical AI development.
• Experience implementing evaluation dashboards and tracking model performance using AI metrics.

Key Skills

Education

Any Graduate