Purpose: Tuberculosis (TB) is one of the world's major infectious diseases, with an estimated 1.4 million case fatality rate in 2020. The increase in the incidence of resistant strains of Mycobacterium tuberculosis (Mtb) to first line drugs has dealt a major blow to TB control. Inadequate and untimely detection of TB infection is a major contributor to the high incidence in low-income countries like Nigeria which shares 50% of the global prevalence with seven other high-burden countries. Genomic technologies combined with computational tools such as machine learning (ML) and artificial intelligence (AI) are promising emerging machineries towards achieving real-time surveillance, prompt and accurate diagnosis of TB. Methods & Materials: We developed a geo-genomic portal of data on TB cases in three pilot states (Oyo, Lagos and Ondo) of south-west Nigeria for real time surveillance and as a data repository for ML and AI research to guide TB management in Nigeria. The portal features patient demographics and geolocation for real-time surveillance of transmission trends, baseline chest X-ray for ML and AI diagnostics algorithms as well as the genome sequence of the Mtb isolates obtained from patients for phylogenetic and ancillary genomic analyses. Results: The portal currently hosts 512 records with 28% (146) positive drug-resistant TB (DR-TB) cases and 72% (366) drug-susceptible TB (DS-TB) cases. There are 101 chest X-ray images from 21 DS-TB and 80 DR-TB cases. Mtb isolates have been obtained for genomic sequencing from 406 patients. The portal is actively maintained, and its size grows as more data becomes available. Conclusion: The web interface of the portal reveals at a glance the location with the highest burden (Alafara with geocode 7.382775, 3.90489) while the dynamics of transmission can be further elucidated using the genomic data. This resource can be a critical aid in national policy formulation for the control of TB.