I manage the technical operations framework for telecom operations — ensuring a smooth transition from the construction phase to full operations, with AI-assisted technical monitoring, AI-driven operational automation, and AI-assisted incident response alongside vendor management, and coordinating with engineering and construction teams on equipment commissioning, system handovers, and operational readiness.
AI is now at the centre of that work. I built a RAN-specific generative-AI assistant on performance and configuration data in Amazon Redshift and took it from proof of concept to pilot: engineers get plain-language answers instead of digging through raw PM/CM data, daily network summaries instead of manual reporting, and likely-cause explanations of KPI movements that cut investigation time. I also replaced a vendor KPI-monitoring subscription with an in-house platform built on AI-driven anomaly detection, automated KPI summaries, and root-cause suggestions. Together, that work delivered roughly $1M in annualized savings. I also supported the business case for 30–50% MVNO bill savings as part of Comcast's wireless economics.
My current interest is where that terrestrial story goes next: Direct-to-Device (D2D) and Non-Terrestrial Networks (NTN) — ordinary phones connecting directly to satellites beyond terrestrial coverage, and what that means for device performance, spectrum, and operations. That is where AI comes in next: AI-assisted performance management for satellite-terrestrial networks, so hybrid coverage is run with the same discipline as the terrestrial network. More on D2D & NTN.
The experience behind the results
Connect on LinkedIn