← All case studies

6× faster vessel dashboards with a new Aggregator service

A Java 21 service built from scratch that turns captains' noon reports and telegrams into daily and voyage-leg aggregates, so key widgets read ready answers.

Result
6× faster response on key vessel-metric widgets
Context
Maritime-AI vessel-monitoring platform, 1,000+ vessels
Role
Designed and built the service from scratch
Stack
Java 21 · Spring Boot · PostgreSQL · AWS
  1. Noon reports & telegramsReported per vessel, per event
  2. AggregatorAggregates by day and by voyage legnew service, Java 21
  3. AggregatesReady-to-serve daily and voyage-leg views
  4. Client widgetsKey vessel-metric widgets6× faster
Figure 1 — Simplified data flow

Situation

Captains’ noon reports and telegrams arrive per vessel, per event. Clients look at the fleet differently: by day and by voyage leg. Key vessel-metric widgets had to bridge that gap at request time, and their response times showed it.

Task

Make the key widgets fast by serving data in the shape clients read it, without making request-time queries heavier. I owned the service from design to production.

Action

  1. A dedicated service instead of heavier queries. I designed and built Aggregator from scratch on Spring Boot and Java 21, owning the data from ingestion to serving.
  2. Aggregate along the axes clients use. It ingests noon reports and telegrams and produces aggregates by day and by voyage leg, so widgets read precomputed answers instead of raw reports.

Result

  • 6× faster response on key vessel-metric widgets, and a smoother experience for clients.

Hiring a Staff or Lead backend engineer?

I'm looking for a fully remote role on a B2B contract, with real ownership of architecture on data-heavy systems.

or write to windgub@gmail.com

Need senior backend help on a project instead? See freelance and contract work.