Software Engineer
I work across the full stack and the full lifecycle: service architecture, backend and UI implementation, AWS infrastructure, production support, the team's AI-assisted engineering tooling — and contributed to Manila office's internal platform on the side. 390 merged merge requests authored; 350+ teammates' merge requests reviewed and approved.
The order engine
- One of the core engineers rebuilding Mantle's order-processing service — the system that confirms an order, calculates pricing and discounts, credits refunds, and expands bulk collections into individual titles — including the performance work that lets a single order carry up to 40,000 titles.
- Built the refund-crediting capability twice: the original version in the legacy platform, including its customer email flows, then its successor in the rebuilt stack.
- Delivered the 2025 product-upsert programme end to end — the backend search and validation services, the cloud functions behind them, and the Vue.js screens on top (progress indicators, confirmation checklists, duplicate-report downloads). Genuinely full-stack: 126 commits in the customer-facing web app across four years.
Finance & fulfilment integrations
- Built the pipelines that push confirmed orders into Cambridge's SAP finance system and the Access to Research platform, plus the two-way integration with EBSCO's MOSAIC ordering platform — order placement, status callbacks, and fulfilment updates.
- Built a brand-new reconciliation service from its first commit — two-thirds of its entire commit history is mine. It keeps Cambridge's internal record of what each customer owes and has paid lined up with SAP's records, with built-in safeguards so it can't overload SAP.
- Added a secure lookup so customer-service staff can pull a library's SAP account and billing history without leaving Mantle.
Access, identity & catalogue
- Built the access-control flow that grants and revokes a library's right to read a purchased digital book or journal, plus the staff tools for finding and repairing missing entitlements.
- Built the bridge to the university's central identity system so a customer's account details stay in sync everywhere, and contributed to the reporting service that produces order summaries and library catalogue records (42 commits).
Cloud infrastructure & reliability
- Designed and built the independent AWS functions powering the order pipeline — confirmation (43% of that repo's commits), entitlement granting, crediting, product explosion — each with automatic retries and instant Slack alerts if something fails silently.
- Built Mantle's job-scheduler service essentially solo (28 of its 29 commits): one place to define the platform's recurring jobs, starting with the ledger sync against SAP.
- Hardened the infrastructure: message queues and safety-net queues defined as code, credentials moved from plaintext into AWS Secrets Manager, deploy pipelines migrated from long-lived AWS keys to short-lived per-deploy credentials (OIDC), and personal data scrubbed from logs across three services.
Production ownership
- Third-line support for the live platform throughout: ~50 investigation and data-repair tasks — missing entitlements, wrong crediting emails, orders stuck mid-pipeline, monthly reporting operations. 220 of my 234 Mantle-board tickets (94%) are delivered and live.
- Own the customer-notification services (117 commits) — the emails libraries receive for order confirmations and credits — from new recipient and content rules to diagnosing why the wrong email went out.
Internal platform — Manila office
- Full-stack engineer on the office's internal workforce platform: the DTR (timekeeping) system — shift scheduling, automated leave harvesting, payroll reporting with real rules (ADP exports, night-differential computation, holiday-duty logic) — and the seat-reservation system, including the AWS Lambda that auto-releases unused seats.
Developer platform & AI-assisted engineering
- Designed and built the team's governance framework for AI-assisted development (Claude Code): a spec-driven pipeline — written proposal, implementation plan, code, verification, each gated by human sign-off — backed by enforcement hooks that can technically block a non-compliant commit, five specialised pipeline agents, and cost-tiered model routing. 210+ commits, a third of the tooling repo's history, grown from a single ~5,200-line foundational commit.
- Audited how the AI loads its reference material and cut it by roughly 47%, making every task cheaper and faster with no loss of enforcement.
- Wrote the onboarding documentation for ~40 services — legacy and rebuilt alike, and maintain the two shared libraries plus the one-command Docker environment that spins up all ~30 services locally. The team's go-to person for this tooling.
TypeScriptNode.jsAwilixFastifyHapi.jsNuxt.jsObjection.jsVue.jsVuetifyVuexKnexMySQLNATSRedisAWS CloudFormationAWS EventBridgeAWS LambdaAWS SQSAWS Secrets ManagerClaude CodeDockerGitLab CI/CDJest
