Website Visitor Analyzer

The project involved developing an application to rank website visitors based on the activities they undertake on a website. The ultimate aim was to be able to identify, with high levels of precision, visitors who were likely to convert. To arrive at this conclusion, a rule-based scoring engine was to be used along with an AI model. Very broadly, Google Analytics was to be used to record all the activities performed by a user on a site. This data was then to be passed through a rule-based engine and an AI model to arrive at a score which was a tentative reflection of the chances of conversion. This outcome can be of very high relevance to companies desiring to improve their website experience, bring more predictability in user segmentation and outreach campaigns.

Client : Confidential

Location: UK

Industry: Media

Python, FastAPI, PyTorch, Pandas, seaborn, NumPy, Vertex AI, Vue.js, Postgres, SQLAlchemy, Docker, Google Cloud Run, Google BigQuery, Google Pub/Sub, Google Cloud Scheduler, Google Cloud Build, Google Secret Manager, Google Container Registry,  Pandas, OKTA (SAML), Terraform (IaC), Jest, pytest