Remote: Yes, that's what I'm looking for. I'm GMT+5, which covers Gulf hours fully and European mornings. I've done partial US overlap before and can again.
Willing to relocate: Yes, if you sponsor the visa. Gulf, Europe, Japan, Singapore all work.
Technologies: C++, Python, TypeScript, SQL, Rust (learning). FastAPI, SQLAlchemy async, Celery, NestJS. PostgreSQL, Redis, Kafka, Elasticsearch. Docker, Linux, Terraform, AWS, GCP. LangChain and RAG plumbing, some PyTorch and OpenCV.
Résumé/CV: https://www.abdulrafayzahid.dev/resume.pdf
Email: sarbzga@gmail.com
Backend and systems engineer, ~3 years. Two things worth pointing at:
I scaled a C++ TLS WebSocket gateway from 200 to 5,000 concurrent clients on a zero-trust network access product, through connection pooling, network I/O, memory work and a threading redesign. Same job, I got the C++ client running on Solaris for a banking client after the task had stalled between a few engineers. The vendor's tunnelling API documentation was wrong, so somebody had to sit down and find out what those calls actually did against the real OS.
At my current job I built an email infrastructure platform on my own, 131K lines in about two months. FastAPI, Postgres, Redis and Celery for the control plane, driving KumoMTA edge nodes across several regions, plus the agent daemon that runs on each node, plus the DNS and deliverability automation (SPF/DMARC/DKIM publishing, warmup cycles under a Redis lock, blacklist monitoring). It runs at 200K+ emails a day across 1,000+ domains and beat the Salesforce tool it replaced on campaign response rates.
The work I like is where something is measurably too slow or too fragile and the fix turns out to be a few layers down, in the connection pool or the event loop or the index. Also happy to talk about Kafka pipelines feeding ML training and inference, which I built for a multi-tenant device profiler.
On logistics: I can invoice directly as a contractor, or go through whatever EOR you already use. Deel, Remote and Oyster all cover Pakistan.