Erythrocytes are a major component of blood, crucial for the proper and efficient functioning of the human body, primarily responsible for transporting gases to tissues and maintaining systemic equilibrium. erythrocytes develop through several stages, and reticulocytes (immature red blood cells) are the penultimate stage of erythroid development that appears in peripheral blood, representing the precursor to mature erythrocytes. reticulocyte count in peripheral blood is often range from 1-2% of total rbcs and can be used to ascertain many disease states such as bone marrow dysfunction, anaemia, pathogen infections (plasmodium, streptococcus, etc), kidney failure, and chronic liver diseases. however, cell counting, especially blood cell counting by image analysis, is done manually, which is tedious, fallible, and biased to the observer counting. as an alternative to the traditional counting methods, automated methods using machine learning revolutionize biomedical data analysis, providing solutions that are faster, cost-effective, and less prone to biases compared to conventional methodologies. here, we have developed a novel convolutional neural network (cnn)-based model optimized for execution on standard cpus, eliminating the need for expensive hardware or sophisticated analytical platforms. our model distinguishes reticulocytes from mature red blood cells in new methylene blue (nmb)--stained blood smears with an accuracy of 90%. we used more than 200 nmb-stained images from leukocyte-depleted blood to train and optimize the model for immature reticulocytes (stained positive with nmb, intensity, and pattern of which depends on the developmental stage of the reticulocyte) and mature rbcs (no staining with nmb). the training performance evaluation metrics demonstrated a mean average precision (map50) of 0.88, a precision of 0.83, a recall of 0.88, and an f1 score of 0.87. our model was able to successfully count reticulocytes with an accuracy of more than 90% from unknown samples, which were subsequently cross-verified through microscopy and counting. given the importance of reticulocyte dynamics in blood and its clinical relevance, the newly developed model is important and easy to adopt biomedical applications that can be achieved on a simple pc.