Architecture is the craft. Performance is the obsession.
Software Architecture · Distributed Systems · Blockchain & DeFi
Cairo, Egypt · LinkedIn · Email
Five years working across backend platforms, identity verification, financial infrastructure, and blockchain systems.
The interesting challenges rarely sit inside a single function. They appear between services, across database boundaries, under concurrent load, or when an external dependency stops behaving as expected.
Service boundaries, data ownership, authentication, authorization, and communication between services.
The difficult part isn't choosing between microservices and a modular monolith. It's deciding which responsibilities belong together, which need isolation, and how the system maintains consistency without unnecessary coupling.
Architecture should make the system easier to understand, change, and operate, not simply introduce more components.
Query plans, indexing, caching, asynchronous processing, concurrency, and profiling.
Most performance improvements start by understanding where time and resources are actually spent. Sometimes that means removing database round trips or reducing unnecessary work. Other times, it means changing an execution path.
Python for backend services and data pipelines. Rust for processing paths where profiling justifies the added complexity.
Distributed systems fail in ways that aren't always obvious.
A request can time out after succeeding. A message can be delivered twice. A worker can restart halfway through a job. An external provider can become unavailable while everything else remains healthy.
Retries, idempotency, timeouts, backpressure, recovery, and observability are part of the design, not additions after deployment.
Solana-based infrastructure for market data, on-chain transaction analysis, and automated trading.
Work includes DEX integrations, liquidity and price aggregation, grid trading, arbitrage engines, and transaction analysis.
The engineering challenges revolve around event freshness, execution latency, transaction construction, large-scale data processing, and the reliability of systems operating against constantly changing market conditions.
Python · Rust · TypeScript · Go
Django · FastAPI · PostgreSQL · Redis · ClickHouse
RabbitMQ · Kafka · Docker · AWS · Linux
Make it work. Understand why it works. Know how it fails.
