Faisal Naveed

- Designed and shipped highly available, horizontally scalable Go services for the cloud, serving 1M+ users across five product lines at 99.9%+ uptime — stateless service design, health probes, graceful shutdown, and autoscaling under variable load. - Applied multi-threaded/concurrent design patterns throughout: goroutine worker pools, channel fan-out/fan-in, context-based cancellation and timeouts, and sync primitives for shared state — holding sub-200ms p95 latency at peak. Enforced -race in CI to catch data races before merge. - Owned production diagnostics and troubleshooting as part of a formal on-call rotation: triaged live outages from Grafana/Prometheus alerts, correlated metrics, logs and traces to isolate root cause under time pressure, and drove written postmortems and follow-up fixes. - Performance tuning as a first-class discipline — profiled CPU, heap, goroutine and block contention with pprof, eliminated goroutine leaks and hot allocation paths, and cut server cost 60% while improving latency, all measured against p95/p99 rather than averages. - Built microservice architectures on REST and gRPC over HTTP/2, replacing chatty REST hops with multiplexed protobuf contracts and reducing inter-service call overhead by 30%. - Ran workloads on Kubernetes (GKE) and Docker — authored manifests, resource limits, liveness/readiness probes, HPA policies, and zero-downtime/blue-green rollout strategies; managed cloud infrastructure as code with Terraform. - Operated across Linux hosts and containers day to day: bash automation for deployment, log rotation, backup and health-check tooling; diagnosed container, networking and disk-pressure issues at the OS level. - Configured load balancers, TLS termination, request routing and health checks on GCP; audited and reclaimed orphaned network resources during an infrastructure review, contributing to $500K+ annual cloud savings. - Engineered event-driven pipelines on Kafka for asynchronous state propagation, decoupling write paths from consumers and sustaining 3M+ monthly transactions without backpressure failure. - Worked across SQL and NoSQL: PostgreSQL schema design, indexing and partitioning that cut query time 70%; MongoDB for high-write document workloads; Redis for caching and rate limiting. - Drove engineering standards in a rapid-delivery agile environment — peer code review, static analysis and coverage gates in Jenkins/GitLab CI, table-driven unit tests plus integration and E2E suites reaching 95% coverage on a production platform. Mentored engineers on Go concurrency. - Collaborated with distributed teams across UK, South Korea and US time zones; presented technical work publicly as a conference speaker and panel moderator.

Pixelpulp

June 2019 - Present

Engineered a high-fan-out aggregation service in Go that queried dozens of unreliable third-party HTTP APIs concurrently — bounded goroutine pools, per-source timeouts, circuit breakers and partial-result semantics held sub-200ms average latency even when upstreams degraded.

Cashero

June 2019 - Present

Built Go microservices for high-concurrency financial operations where correctness under concurrent access was non-negotiable — idempotency keys, transactional boundaries and careful lock/contention design.

BBGSwap

June 2019 - Present

Built a latency-sensitive real-time backend in Go consuming live market feeds and order flow over WebSockets — the hardest problems were concurrency and backpressure: fan-out to thousands of concurrent subscribers without blocking the ingest path, and shedding load gracefully under burst.

MyB

June 2019 - Present

Led the Go backend for a platform that scaled past 1M in-region users, applying domain-driven boundaries to keep services independently deployable as the system grew.