Daniyal A.

Most backends do not fail on the happy path. They fail when a connection drops mid request, a job dies halfway through, or a webhook fires twice and nobody notices for a week. I build Go services that hold up in those cases. Most of my production work has been EV charging and IoT infrastructure, which is a good proving ground because the devices are unreliable by nature and the state has to stay correct anyway. - OCPP 1.6J central system in Go. WebSocket charger connectivity, Redis runtime state, AMQP command dispatch, Kinesis telemetry, session recovery after restarts. - Telemetry pipeline: Kinesis to DynamoDB to S3 Parquet to Postgres, with stale session finalization and tariff based cost calculation. - Go REST services: auth, webhooks, rate limiting, background workers, queue recovery, Postgres schema design. Also full stack product work on an AI app builder platform (Next.js, TypeScript, Python, AWS ECS/Fargate) If something in your system is unreliable and you are not sure why yet, tell me what it does when it breaks and I will give you a straight read on whether I can help.