Mahaboob Sharif Bikarisabgari

Having 7+ years of recent experience as a Data Scientist, Machine Learning and Big data architect.

Overall 17+ years of experience in software design and development.

Good knowledge and hands on experience of deep learning and machine learning models in time series and NLP applications.

Expertise in anomaly detection and forecasting on time series data using Machine Learning and Deep Learning techniques.

Knowledge and expertise on handling, analyzing , ETL processing and Machine Learning implementation for the 3G,4G mobility and network security data.

Worked as BitBucket/GIT/GIT Hub contributor as part of my role.

Experience of software development in agile development methodologies.

9 +years of previous experience in Internetworking, Telecom and datacom experience in QoS , L2,L3 protocols software development for Cisco routers.

Developing automation script using Python ,Shell scripts.

1. Support Cost optimization for Network Devices: NLP, Machine Learning based solution.

· The cost to support the cisco network devices widely deployed in access and aggregate data networks is high and increasing year on year due to high customer ticket inflow, High replacement of devices and high-ticket resolution times. To avoid this high support cost, An artificial intelligence based solution for Ticket investigation is developed based on Machine Learning and Deep Learning techniques.

· Design and Implementation of software for text summarization on the ticket information and the mail chains available as part of ticket investigation.

· Classifying the historical data into suitable clusters and segregating them as a pre-classification to Machine Learning and deep learning models.

· Identification of Suitable patterns from the logs and captures attached to the tickets.

· Design and implementation of data mining, NLP techniques to identify the patterns from the logs and captures.

· Design and implementation of suitable machine learning models and deep learning models for classification of the customer tickets and identifying the similar SR’s.

· Identifying the cause of issue based on pattern’s in logs collected.

2. Machine Learning based Matrix Subscriber Threat Intelligence Platform for network security of a mobile service provider.

· This solution is to develop machine learning based solution to provide insight of security incidents and risk in the service provider mobile network using the data from network firewalls, Threat intelligence data (Thread grid ,MISP) and subscriber control plane and user plane data.

· High Level design for Matrix subscriber threat intelligence platform for AT&T Mexico.

· Design and implementation of ML based prediction of IoC(Indication of Compromise) for security events with classification of priorities.

· Design and implementation of ML based baseline model creation.

· Design of Threat intelligence map(Geo representation of subscriber malicious activity) for Mexico representing country level, state level, city level, cell-tower level threat analytics along with ML based risk assessment.

3. Customer Experience Management (CEM) and Customer Care (CC) solution for mobility service provider.

· The overall project is about design and implementation of Matrix Subscriber data analytics Platform. The solution comprises of data engineering pipeline to processes the control plane & user plane data and Web UI applications for CEM/CC application. This project design and implements end-to-end processing of probe data through pipeline structure where each stage of data processing will be covered (e.g., enrichment, cause code, fact tables, counter extraction, KPI/KQI etc).

· Design and implementation of Machine learning based customer segmentation implementation.

· Design and implementation of anomaly detection and forecasting algorithms on time series data of the customer using ML and DL techniques.

· Design and implementation of etl processing enrichments for various user-plane and control plane interfaces like s1mme, s11, s6a, sgs, iucs, iups,isup,map,camel,gn,sip and userplane.

· Design and implementation of creating different aggregation tables for various dimensions like subscriber, device, cell and core.

· Design of Creation of control plane and userplane KPI/KQI tables from the enriched data, counter tables , cause code tables and aggregation tables.

· Design and implementation of Airflow Dags for enrichment and aggregation stages and creating machine learning pipelines.