Prompt Engineer

Role Description: LLM Prompt Engineer
Location: Basking Ridge NJ

About the Role
We are seeking a highly creative and technically strong Prompt Engineer to design, optimize, and scale LLM-driven solutions within an Agentic POD architecture. The ideal candidate will have hands-on experience in LLM optimization (Gemini, Claude, Copilot), prompt engineering patterns, and Retrieval-Augmented Generation (RAG), along with a focus on improving engineering throughput and automation.

Role and responsibilities

1. Design, develop, and optimize prompts and prompt chains for LLM-based applications
2. Work within Agentic POD architectures to enable seamless interaction between AI agents
3. Optimize performance of LLMs such as Gemini, Claude, GitHub Copilot, and other foundation models
4. Develop and implement prompt patterns (few-shot, chain-of-thought, ReAct, tool-augmented prompting)
5. Build and enhance RAG (Retrieval-Augmented Generation) pipelines for accurate and context-aware responses
6. Create reusable prompt templates and frameworks for enterprise-scale applications
7. Improve engineering throughput by leveraging AI-assisted development workflows
8. Evaluate and fine-tune prompts using LLM evaluation frameworks and metrics (accuracy, latency, cost)
9. Collaborate with AI engineers, data scientists, and product teams to deliver optimized solutions
10. Ensure responsible AI practices, bias mitigation, and prompt safety controls
11. Continuously experiment with new prompting techniques and LLM capabilities

Technical skills requirements

The candidate must demonstrate proficiency in,

· Strong understanding of LLMs and prompt engineering techniques
· Hands-on experience with LLM optimization (Gemini, Claude, Copilot, OpenAI models, etc.)
· Expertise in prompt patterns:
· Few-shot / zero-shot prompting
· Chain-of-thought reasoning
· ReAct / tool-based prompting
· Experience with RAG architectures (retrievers, embeddings, vector databases)
· Ability to design scalable prompt templates and reusable frameworks
· Strong analytical skills for evaluating and improving model outputs
· Familiarity with Python for prototyping and integration

Nice-to-have skills
· Experience with LangChain / LangGraph / AutoGen
· Familiarity with vector databases (FAISS, Pinecone, Weaviate)
· Knowledge of cloud platforms (GCP preferred – Vertex AI, AI Studio)
· Understanding of workflow orchestration (Airflow / Cloud Composer)

Qualifications
• Overall 4 + years with 2-7 years of relevant work experience in Agentic AI Development

B.Tech., M.Tech. or MCA degree from a reputed university