ABSTRACT Introduction: Neonatal sepsis is a major cause of neonatal mortality, the clinical outcome of which depends on early diagnosis and initiation of appropriate antibiotics. The emergence of multi-drug resistant strains has limited the choice of available antibiotics. Thus, antibiotic resistance pattern of pathogens is critical for both therapy and infection control. Materials and Methods: This observational study was conducted in the Microbiology department of Tertiary care level hospital affiliated with medical college. Venous blood collected aseptically before initiation of antibiotic therapy. For neonates 2 ml of blood & for older children 5 ml of blood was collected from peripheral vein with aseptic precaution. The blood culture bottles were sent to laboratory, where they incubated in BACTEC (FX40) if bottle was positive then it sub-culture were made on MacConkey & Blood agar & incubated in appropriate temperature & further identification was done according to standard guidelines.Antimicrobial susceptibility testing was done by Kirby-Bauer disc diffusion method and interpretation was done according to CLSI guidelines. Results: Among 100 blood samples processed from clinically suspected neonatal septicemia cases, 28 (28 %) samples showed growth.Among 28 isolates, 15 were from early onset septicemia (EOS) and 13 were from Late Onset Septicemia (LOS). Among 28 culture positives, 20 (71.4%) were from males and 8(28.6%) were from females, thus showing a male preponderance.Gram-positive bacteria were responsible for most cases of neonatal sepsis. Coagulase negative staphylococci (CoNS) were the most frequent isolated pathogens in EOS and LOS, followed by Klebsiella pneumoniae, Staphylococcus aureus, Enterococcus spp and Escherichia coli.The antibiotic sensitivity showed that most of the Gram negative bacteria were highly resistant to the commonly used antibiotics like Ampicillin and Gentamicin. Conclusion: Neonatal sepsis is a medical emergency and empirically treatment is started without microbiology report. If local microbiological databases are available with information regarding the commonly isolated organisms and their drug resistance patterns, it can help the clinicians in planning of therapy which in turn reduces neonatal mortality and morbidity.