Olavo Wilke

MSE searches, scores, and orders parts across multiple suppliers while staying in sync with internal inventory and finance. It fans out parallel vendor queries, short-circuits the moment internal stock satisfies demand, and ranks the rest by price, delivery time, and reliability. Orders persist instantly and fulfill asynchronously, so vendor slowness never blocks the customer.

• Spring microservices with a central orchestrator/API gateway, PostgreSQL as system of record, and Amazon SQS for asynchronous fulfillment.
• Each vendor wrapped in a dedicated adapter (anti-corruption layer) handling auth, retries, and errors — independent scaling and reduced blast radius.
• Parallel vendor search with early termination when internal inventory satisfies demand; scoring by price, delivery time, and reliability.
• Async order flow: persist as PENDING → publish to SQS → vendors fulfill independently — zero vendor latency on the critical path.
• NetSuite integration for financial sync and persistent audit trails for compliance.

• Spring
• Microservices
• PostgreSQL
• Amazon SQS
• NetSuite
• AWS

The core Java API processes every bet and win in real time, so the constraint was simple: tournaments could not slow it down. I wrapped the core API with non-blocking RabbitMQ connectivity, fanning out every bet and win to a new service — gator-messaging — that ran all tournament logic off the critical path, with live leaderboards served straight to the back office.

• Wrapped the latency-sensitive bet/win API with non-blocking RabbitMQ connectivity — tournament processing never touched the critical path.
• Built gator-messaging: a new service consuming bet/win events to run scoring, standings, and prize handling independently of the core API.
• Live leaderboards computed and persisted in a dedicated MySQL table, exposed through a back-office endpoint and updated fully live.
• Full-stack back-office tournament management: one operator, many game providers and players, a configurable prize set, and a date range.
• Drove the discovery and infra work to introduce RabbitMQ — the platform had no message broker before.

• Java 7 / 17
• Spring Boot
• RabbitMQ
• MySQL
• JSP
• Docker
• Jenkins

For one of the largest marketplaces for sports jerseys, hats, collectibles, and fan merchandise, I enhanced a marketing engine that emails users based on the outcomes of the matches they care about. Go routines read registered game events from MySQL and trigger campaigns through Adobe Journey Optimizer, while an event-driven pipeline keeps game state current and Slack keeps operations in the loop.

• Event-driven game state: MDM → Kafka → the klio Go service → MySQL, keeping game status current for campaign triggering.
• Built REST APIs letting back-office systems register 'game events' that control which games participate in a campaign.
• 1-minute routine gathers game updates and triggers Adobe Journey Optimizer campaigns; a 5-minute routine retries failed transactions for reliability.
• Adobe Journey Optimizer integration delivering results-driven emails to audiences in the hundreds of thousands.
• Internal Slack notifications so operations can manage which game events are being processed live.

• Go
• Adobe Journey Optimizer
• Kafka
• MySQL
• REST
• Slack