I will build or fix cdc pipelines with debezium, kafka and postgres

Pakistan

I speak English

181 orders completed

Senior Data and AI Engineer, RAG, LangChain, Snowflake, dbt, Agents, Clickhouse

I build data platforms and LLM pipelines that run in production. 10 years in software and data engineering, since 2016. Data Engineer II at Careem (Uber), where our DataOps team ran 500+ streaming an...

Level 2

Has met high performance criteria and has a proven track record for meeting client expectations.

About this Gig
  • Change data capture breaks quietly. Replication slots get stuck and fill the disk, large columns arrive as placeholders, and MERGE statements overwrite good rows while every dashboard stays green.

I build CDC that holds up, or fix the one you have:

  • Log-based CDC from Postgres or MySQL with Debezium, Kafka, Meltano or Maxwell
  • Loads into Snowflake, BigQuery or Postgres with correct MERGE logic for deletes and updates
  • Fixes for Postgres TOAST placeholder overwrites
  • Replication slot monitoring and a safe recovery path, with no full reload
  • Reconciliation checks so source and target counts and values match

Track record:

  • Rebuilt Postgres replication with log-based CDC into Snowflake for a US delivery-tech unicorn
  • Built an agent that clears stuck replication slots and runs a clean partial refresh, with no one paged
  • Careem: real-time CDC through Maxwell, Kafka and Spark Streaming
  • Wrote up the Debezium TOAST silent-overwrite bug and its three-layer fix

Langugae:

English

Technical expertise:

Apache Airflow

•

Dagster

•

Apache Spark

Expertise:

Data Pipelines

•

Streaming Data Processing

Industry:

Audio & video

•

Cyber security

•

Data analytics

My Portfolio

Other Data Engineering Services I Offer

Related tags