Fresh prospect data, on demand
FunnelBeam’s sales-team users needed real-time access to enriched prospect profiles. Behind them, a data pipeline had to ingest company intelligence from multiple sources, enrich it and serve it through search indices and several delivery channels, at high throughput and at scale.
How the data moves
We built the platform as microservices in Python, Node.js and Go:
- Apache Airflow orchestrates the multi-source ingestion, enrichment and ETL workflows
- Kafka carries real-time event streams between services
- Elasticsearch provides fast full-text search across millions of company records
- A Django backend and Flask APIs serve specific enrichment services
- A React portal puts the intelligence in front of sales teams
Infrastructure is defined as code in Ansible and CloudFormation, with Dev, Demo, Vagrant and Production environments.
Scaling the parts separately
Each service scales independently with its own load, and the Airflow pipelines keep the data fresh, so sales teams work from current, enriched prospect intelligence.