🤖 Building a Telegram Bot with Apache Kafka, Go & ksqlDB
summary
A Telegram bot reads Bradford council carpark CSVs every 10s (piped with kafkacat) and writes events to Kafka; ksqlDB parses the CSV, converts timestamps, writes protobuf-serialized events, and builds a materialized view table aggregated by car park name (latest timestamp, capacity, average empty spaces). The Go bot issues pull queries for lookups and spins push queries for threshold alerts (example threshold 500, demo alerts fire — e.g. 591 or 1100 spaces). Kafka Connect sinks the cleaned stream to Elasticsearch/Kibana for maps and dashboards; demo runs from docker-compose.
AI-written summary · about
chapters (6)
- 0:00Intro
- 1:07Where's my nearest carpark with available spaces?
- 2:46Demo
- 30:23What are the key pieces of the design?
- 35:45Why build it this way?
- 36:51Outro
related talks
22mBuilding a Telegram bot with Apache Kafka, Go, and ksqlDB.
29mGo Big with Apache Kafka
34mETL on Kafka + Confluent: Problems and Solutions with Go
28mManaging and Scaling Real Time Data Pipelines using Go
05mGo for Real Time Streaming Architectures
25mBeyond Kafka: Leveraging Event-Driven Architectures with Go and NATS JetStream
23mImplementing Event-driven Systems with Go
54mComplex Concurrency Patterns with Go