The volume of data has grown on a huge scale in recent days. As a result, there is always a plethora of choices in any product or service. It is very natural to get lost in the sheer amount of such choices and finding hard to make decisions. This project aims at addressing this problem by using recommendation system. This project makes use of Apache Spark, a distributed big data processing framework. Compared to the traditional map-reduce paradigm, Spark offers a huge advantage in the way it handles iterative and interactive algorithms. The dataset for the project were obtained from Stanford Network Analysis Platform and we observed that the Alternating Least Squares algorithm built into the MLlib library performs well by making use of the Spark RDDs for fast execution of iterative algorithms. We also observed that the results of the recommendation engine are accurate.