The husbandry is very important business of Indian farmer. Dairies collect milk from farmer everyday & payments for this milk are done according to the quantity and quality of milk. This rate depends on various quality parameters of the milk. Fat is one of most important component of milk. Due to its higher importance along other quality parameters of milk, it becomes mandatory for the determination of fat percentage especially for the milk vendors, small dairy owners etc. The classical method used to determine the fat percentage in milk is Gerber method. It’s a chemical test which is destructive in nature. Due to inclusions of various chemicals it becomes time consuming procedure for determination of fat in milk. Due to this, the number of samples that can be tested is very low in a given period of time. This paper presents the design and development of Milk Fat Analyser. The system is built by interfacing Light Dependent Resistor (LDR) with Raspberry pi Microprocessor. The parameter that is measured is fat percentage. The result obtained is displayed on screen which provides a good interface to the user. This project aims to provide a low cost, fairly accurate and reliable system to the farmers, small dairy shop owners and similar small factories where quality of milk matters while processing it to make other milk products and also for supplying milk consumed by people at large. Experimental results show that the method with Machine learning algorithm can detect fat with fair accuracy.