India is one of the most suitable countries for agriculture, yielding numerous varieties of crops annually. However, in recent days, the condition for agriculture has changed drastically due to the influence of various environmental factors like climatic conditions, nature of soil, pollution levels, natural hazards etc. Addressing these challenges in Indian agriculture, a portable crop prediction system equipped with various sensors is developed for real-time data collection on temperature, humidity, pH, soil fertility, soil moisture, and rainfall. This system leverages machine learning and IOT technology, integrating historical and real-time sensor data collected to accurately predict crucial environmental factors. By implementing MPL classifier, XG Boosting classifier and bagging classifier machine learning algorithms the datasets are trained and tested for its accuracy, there by finding the most suitable machine learning algorithm for crop prediction. XG Boosting classifier is the most suitable machine learning algorithm for crop prediction. Driven by Arduino Mega and Zigbee modules, the system utilizes the XG Boost classifier algorithm to predict crop yield and recommend suitable crops. The system aims to enhance agricultural productivity by providing farmers with live sensor data display and crop recommendations through a web-based platform, facilitated by Node MCU. This Portable crop prediction system provides farmers with actionable insights, enabling them to adapt to changing environmental conditions and optimize crop productivity. It also offers farmers comprehensive environmental profile of their agricultural land, enabling informed decision-making and optimization of agricultural practices. This promotes sustainable farming practices and efficiency, resonating with the environmentally conscious mindset of Gen Z farmers.