ABSTRACT Purpose To evaluate the current communication skills of resident physicians and identify areas for improvement. Study design Using a cross-sectional design, data from medical students, residents, fellows, and faculty were assimilated at one of the largest academic medical centers in a low-and-middle-income country via a self-administered, validated survey with a 5-point LIKERT scale. One-way ANOVA was used to compare intra-group means, followed by the Bonferroni test, with a p-value <0.05 considered significant. Heat maps using means were generated and color-coded to signify the level of communication skills expertise. Results A total of 320 participants (119 residents, 34 fellows, 78 medical students and 89 attendings) completed the survey. Most residents (58%) reported not having received formal communication skills training. Major barriers impeding effective patient-resident communication were long working hours, inadequate time, and difficulty understanding patients’ language (p < 0.001). Residents’ self-evaluation of communication skills was significantly higher than observer evaluations from faculty, fellows, and students, reflecting a lack of self-awareness and overestimation of communication expertise. Observer cohorts rated residents moderately on para-verbal communication skills while giving the lowest ratings for breaking bad news (mean 2.8, p-value <0.001) and residents’ experience in dealing with workplace conflicts (mean 2.8, p-value = 0.023). Conclusions Our study identified a clear and critical need for a structured, longitudinal, competency-based communication skills program for residents. Targeted efforts towards bridging the gaps in communication skills identified in this study can improve trainee communication skills expertise and augment patient satisfaction rates to improve the quality of care in developing countries.