Generative AI Engineer

We’re looking for a highly skilled and hands-on RAG (Retrieval-Augmented Generation) & Prompt Engineer to join our applied AI team. You’ll work with cutting-edge open-source and proprietary LLMs (like LLaMA, Mistral, Claude, GPT-4o, etc.) to build, prompt, and orchestrate intelligent agents that are capable, reliable, and production-ready.

This role is perfect for someone who has experience developing prompt chains, implementing tool-calling workflows, and debugging AI agents at scale.

Key Responsibilities

• Design, develop, and iterate on prompt strategies tailored to downloadable models and major APIs (LLaMA, Mistral, Claude, GPT-4o, etc.).
  
• Architect and implement RAG pipelines with a deep understanding of embedding models, retrievers, and context optimization techniques.
  
• Create prompt chains and tool-calling workflows for dynamic agent behavior using Responses API and similar frameworks.
  
• Design, test, and deploy foolproof agent architectures using OpenAI tool calling and agent protocol layers.
  
• Write robust Guardrails and control flows for agents to prevent unintended behaviors and ensure task compliance.
  
• Debug and maintain agent codebases, ensuring reliability and scalability of deployed services.
  
• Apply basic knowledge of OpenAI Operator and related orchestration tools to manage agent lifecycle.
  
• Collaborate with researchers and infra teams to optimize prompt efficiency and latency.
  

Must-Have Qualifications

• 3 - 5 years of experience in AI engineering, prompt engineering, or applied ML roles.
  
• Proven experience working with both downloadable open-source models and hosted APIs.
  
• Strong knowledge of LLM prompt design patterns, prompt chaining, and failure handling.
  
• Ability to build agent systems that are secure, auditable, and self-healing.
  
• Good coding and debugging skills in Python (or relevant stack) with focus on AI orchestration.
  
• Familiarity with agent deployment pipelines, containerized environments, and CI/CD flows.
  

Tech Stack We Use

• Python, FastAPI, LangChain / LlamaIndex.
  
• OpenAI, Anthropic, HuggingFace.
  
• Vector DBs (Weaviate, Pinecone, Qdrant).
  
• Responses API, OpenAI Operator, A2A SDK.
  
• Docker, GitHub Actions, GCP/AWS.
  

Bonus (Nice-to-Have Skills)

• Experience building agents from scratch, especially with agent transfer logic and persistent memory.
  
• Understanding of Model Context Protocols and how to integrate them into multi-agent LLM stacks.
  
• Familiarity with A2A SDK for agent-to-agent communication and delegation.
  
• Hands-on experience with LoRA / QLoRA techniques for fine-tuning GPT-style models on downstream or domain-specific tasks.
  
• Experience with vector DBs, context compression, or multi-turn reasoning at scale.
  

#LI-SC1  #LI-Hybrid