Prompt Engineering

Job Description- A prompt engineer designs, tests, and optimizes instructions (prompts) for large language models (LLMs) to ensure the AI generates accurate, safe, and useful outputs. Role Overview Primary Goal: Bridge the gap between human business needs and artificial intelligence capabilities. Core Function: Craft system instructions, few-shot examples, and context guidelines, so AI tools produce reliable responses instead of hallucinations (false or made-up information). Key Responsibilities Design & Optimization: Write and refine prompts for chatbots, virtual copilots, and retrieval-augmented generation (RAG) systems. Support context engineering, prompt engineering, and RAG workflows, including chunking, embeddings, semantic search, and knowledge graphs. Help build agentic workflows and AI agents with frameworks such as Google ADK, LangGraph, or CrewAI. Connect agents with external tools/data via MCP; A2A exposure is a plus. Assist with deployment, monitoring, data preprocessing, API development, and code reviews Testing & Evaluation: Test model outputs against accuracy, safety, latency (response delay), and token cost metrics. Library Management: Build and maintain reusable prompt templates and governance standards for enterprise teams. Adversarial Testing: Run stress tests to check for prompt injections (tricking the AI into breaking rules), data leaks, and bias. Cross-Functional Collaboration: Partner with data scientists, software developers, product managers, and business stakeholders. Common Qualifications & Skills AI & LLM Familiarity: Understanding of tokenization, context windows, and APIs (such as OpenAI, Anthropic, or Azure OpenAI). Python for GenAI, data preprocessing, and scripting. Core GenAI concepts: foundation models, LLMs, tokenization, embeddings, and context windows. Hands-on prompt/context engineering and practical RAG experience. Working knowledge of knowledge graphs and Graph RAG. Exposure to agentic AI, tool-using agents, multi-step workflows, and frameworks such as ADK, LangGraph, CrewAI, or the OpenAI Agents SDK. Awareness of tool/function calling, MCP, A2A, agent harness, memory, and guardrails. Experience with GenAI APIs such as OpenAI, Gemini, and Claude; orchestration frameworks such as LangChain or LlamaIndex; Docker, Git, and AI compliance/privacy principles. Communication & Linguistics: Strong command of syntax, writing, and clear logical phrasing. Technical Skills: Varying by role-some positions focus purely on writing and psychology, while technical roles require basic Python, SQL, or software workflow knowledge.