Marcus Wells Back-End Engineer
Seattle, WA • backend@gmail.com • +1 6666-2222
Profile Summary
- Back-End Engineer with 7 years of experience designing and operating large-scale services across e-commerce platforms, developer tools, and B2B SaaS, specializing in distributed systems, API contract design, and high-throughput data persistence.
- Solid technical background across languages (Go, Python, Java), frameworks (gRPC, FastAPI, Spring Boot), data stores (Postgres, DynamoDB, Redis), messaging (Kafka, RabbitMQ), and cloud (AWS, GCP) with strong fundamentals in SQL and Linux internals.
- Deep expertise in domain-driven design, event-sourced architectures, idempotent processing, and zero-downtime migrations, leveraging methodologies such as hexagonal architecture and CQRS to drive maintainable, scalable, and observable services.
- Engaged collaborator working cross-functionally with Product, Mobile, and Platform teams in Agile environments, contributing to API contract reviews, RFC discussions, and on-call retrospectives with a pragmatic, ownership-first mindset.
- Emerging leader who shares technical excellence and fosters a culture of operational rigor and code quality through PR reviews and runbooks, while leading backend guild sessions and authoring widely adopted service templates.
Technical Skills
- Languages & Scripting:
- Go, Python, Java, SQL, Bash
- Frameworks & APIs:
- gRPC, FastAPI, Spring Boot, GraphQL, REST
- Databases:
- PostgreSQL, MySQL, DynamoDB, Redis, MongoDB
- Messaging & Async:
- Kafka, RabbitMQ, SQS, Pub/Sub, Celery
- Performance & Caching:
- Redis, Memcached, CDN, connection pooling
- Cloud Platforms:
- AWS (ECS, Lambda, RDS, S3, IAM), GCP (GKE, Cloud SQL)
- Testing & Quality:
- PyTest, JUnit, Postman, Pact, k6
- DevOps & CI/CD:
- Docker, Kubernetes, Terraform, GitHub Actions, Datadog
Education
Work Experience
- Owned the service architecture for the GitHub Issues platform supporting 120M+ MAU, leading end-to-end design and operation across API contracts, read/write throughput, and on-call reliability within a polyglot service mesh.
- Designed and shipped a GraphQL API for issue search and triage, defining schema versioning policies, deprecation conventions, and persisted-query support, reducing client payload size by 42% and cutting average query latency from 380ms to 120ms across 8 internal consumers.
- Architected a partitioned PostgreSQL schema for issue activity history, using range partitioning by month, covering indexes, and a logical-replication-based zero-downtime migration, reducing p99 query time on 2B+ rows from 2.4s to 180ms.
- Built an event-driven notification pipeline in Go using Kafka with at-least-once delivery, idempotent consumers, and exactly-once dedup via consumer-side keys, processing 400M+ events/day with 99.99% delivery success and sub-second end-to-end fan-out.
- Optimized hot-path API endpoints through Redis read-through caching, connection-pool tuning, and query plan analysis with EXPLAIN ANALYZE, lifting throughput by 3.5x (from 2k RPS to 7k RPS) and cutting p95 latency by 58% during peak load.
- Instrumented services with structured JSON logging, OpenTelemetry traces, RED dashboards in Grafana, and SLO-based alerting via Prometheus, reducing mean time to detect from 18 minutes to 90 seconds and time to resolution by 62% across 6 incident drills.
- Refactored the billing-events domain using hexagonal architecture with adapter-port boundaries, command handlers, and aggregate-root invariants, eliminating 15 long-standing race conditions and shipping the largest domain model rewrite in 4 years without a customer-visible regression.
- Implemented OAuth 2.1 + PKCE authentication for the public Jira API, building scoped JWT tokens, refresh-token rotation, token introspection, and rate-limited brute-force protection, hardening the auth surface against OWASP Top 10 risks and powering 5,000+ active integrations.
- Established the team's testing strategy spanning unit tests in JUnit, integration tests against ephemeral Docker stacks, and contract tests via Pact, raising overall service test coverage from 47% to 88% and catching 22+ regressions during pre-release cycles.
- Owned CI/CD pipelines for 8 backend services, automating canary rollouts on Kubernetes, schema-migration gating, and post-deploy smoke checks, reducing deploy lead time from 4 hours to 22 minutes and rolling back 3 silent regressions within minutes.
- Worked closely with Product, Mobile, and SRE teams across 3 product surfaces to negotiate API contract changes, deprecation timelines, and migration paths, authoring 15+ technical RFCs that shaped the org's backwards-compatibility policy and onboarding 9 new engineers.