Prompt engineering is one of the main parts which utilizes the power and capacity of large language models ie., LLMs efficiently. Large Language models (LLMs) are generally used to generate human-like text, summarization, solving the problems in various fields, understanding the language and translating the language and so on. The potential of LLMs is utilized by creation of effective prompt which is called as prompt engineering through the inputs are given properly. AI models are used for collecting the relevant information about the query. Extracting the relevant responses from various artificial intelligent models by almost all types of people from researchers to school going children. The challenge lies in crafting prompts which reduce ambiguity and give proper direction to the LLM for getting the desired responses. When we concentrate on prompts by adding the important words or using some key words, we can show better results which reflects the role of prompting techniques clearly. A technical document can be prepared by using a few short prompting techniques and creative writing and storytelling can be done effectively by using some key words in the prompts itself. LLMs such as chatGPT3, chatGPT3.5, chatGPT4, Gemini and other models are trained on huge volumes of data which can produce and generate human-like text easily. The conditional prompts allow the users to use some specific keys for extracting the information on the iterative refinement process can also be used to extract information from prompt engineering. The quality of LLM results is evaluated by using relevance, coherence, creativity and specificity. This work explores the strategies and methods of prompt engineering that could enhance the performance and reliability of the LLMs such as few-shot prompting, role assignment and prompt chaining. Effective prompt engineering is the foremost technique to maximize the utility of large language models in various applications. Advanced techniques such as control tokens and multimodal prompts that combine the text with other modalities such as images for optimizing the results of prompt engineering. Retrieval Augmented Generation gets queries from prompts and try to get relevant information from various sources such as search engines or knowledge graphs. Hencs, RAG extends the LLMs by incorporating external knowledge for enriching the model’s responses. The most popular prompt engineering approaches are CoT, ToT, self-consistency and reflection played a major role. Prompt design and engineering are critical and the innovation in the Automatic Prompt Engineering (APE) would dominate in the near future.  This work explores the effective utilization of Large Language Models for creating effective prompts for optimizing the responses so that we can solve complex problems easily and can reach better results in a stipulated time.