Prompt Engineer

Virtusa; New York (NY); On-site; USD 150,000 - 190,000; Full time

Job summary

Virtusa in New York, NY is seeking an experienced prompt engineering specialist to design and optimize prompts for large language models, including GPT and Claude. You will help build robust prompt/agent workflows, leveraging LangChain, RAG systems, and strong Python/API skills.

The role emphasizes evaluation, experimentation, and security practices to prevent jailbreaks and bias, while collaborating with product, engineering, and data science teams to translate business needs into scalable AI

Qualifications

• Prompt design & optimization for LLMs to maximize accuracy, relevance, and consistency across use cases (chatbots, agents, RAG systems).
• Design evaluation frameworks (A/B testing, benchmarking, rubric-based scoring) to measure prompt performance and iterate.
• Deep understanding of LLM behavior (GPT, Claude, etc.), context windows, tokenization, and frameworks like LangChain or orchestration tools.

Responsibilities

• Cross-functional Collaboration – Work with product, engineering, and data science teams to translate business requirements into prompt/agent workflows.
• Technical & Analytical Skills – Strong Python scripting, API integration, data analysis, and prompt safety practices (guardrails, bias mitigation).

Skills

Prompt Engineering

LLMs (GPT/Claude)

Prompt Evaluation & Optimization

LangChain

RAG & AI Agents

Python

API Integration

LLM Security & Guardrails

Job description

Key Skills: Prompt Engineering, LLMs (GPT/Claude), Prompt Evaluation & Optimization, LangChain, RAG & AI Agents, Python, API Integration, LLM Security & Guardrails

Having 8+ years of experience on the following:

Prompt Design & Optimization – Craft, test, and refine prompts for LLMs to maximize accuracy, relevance, and consistency across use cases (chatbots, agents, RAG systems).

Evaluation & Experimentation – Design evaluation frameworks (A/B testing, benchmarking, rubric-based scoring) to measure prompt performance and iterate systematically.

LLM & Tooling Expertise – Deep understanding of LLM behavior (GPT, Claude, etc.), context windows, tokenization, fine-tuning vs. prompting trade-offs, and frameworks like LangChain or similar orchestration tools.

Cross-functional Collaboration – Work closely with product, engineering, and data science teams to translate business requirements into effective prompt/agent workflows.

Technical & Analytical Skills – Strong Python scripting, API integration, data analysis, and familiarity with prompt security/safety practices (jailbreak prevention, bias mitigation, guardrails).