Make rules explicit
Business language becomes readable domain logic instead of hidden assumptions.
Java Backend Developer · Halifax, Canada
Backend systems for work where correctness matters.
I’m Steve Armstrong. I build Java services that turn complex business rules into dependable software—from pension calculations to enterprise APIs and large-scale financial systems.
DOMAINPension calculations
COREJava · Spring · APIs
APPROACHTested by design
01 / CURRENT WORK
Complex rules.
Clear outcomes.
PENSION CALCULATION SYSTEMS
Pension software sits at the intersection of financial rules, long-lived data, and outcomes that matter to real people. My role is to make that complexity behave predictably in Java.
INPUTPlan provisionsEligibility, service, earnings, dates
ENGINEJava calculation logicExplicit rules, edge cases, tests
OUTCOMEReliable benefitsTraceable, reviewable, correct
Test the boundaries
Happy paths are only the beginning; dates, thresholds, and edge cases deserve first-class tests.
Leave a clear trail
Code reviews, test cases, and technical reasoning make sensitive calculations easier to trust.
02 / BACKEND PRACTICE
Built for the whole path to production.
Domain logic
Turn dense financial and regulatory rules into code that is understandable, testable, and safe to change.
- Java
- Business rules
- Financial systems
Backend systems
Design and evolve service boundaries, REST APIs, integrations, and asynchronous workflows.
- Spring Boot
- REST
- Kafka
Data & persistence
Model application data deliberately and diagnose behaviour across relational and document databases.
- PostgreSQL
- MySQL
- MongoDB
Delivery confidence
Build quality into the path to production with focused tests, observability, and repeatable pipelines.
- JUnit
- Testcontainers
- GitHub Actions
03 / FEATURED BUILD
LedgerGuard.
Events reconciled in real time.
A production-shaped Java and Kafka Streams project that turns independent payment and ledger events into auditable financial outcomes.
LIVE ON RAILWAY · CI VERIFIED · OPEN SOURCE
REAL-TIME RECONCILIATION PLATFORM
Designed for the failure modes that make event-driven finance difficult.
LedgerGuard correlates events from two independently produced streams, tolerates out-of-order arrival, detects duplicates, and routes mismatches or missing counterparts for review. The data is synthetic; the architecture and engineering trade-offs are real.
- Java 21
- Spring Boot 4
- Kafka Streams
- Testcontainers
- Prometheus
01 / INPUTPayments + ledger entriesIndependent JSON event streams
02 / BROKERApache KafkaPartitioned by transaction ID
03 / ENGINEStateful reconciliationJoin · classify · deduplicate
04 / OUTPUTOutcome + exception topicsQueryable result projection
7explicit reconciliation outcomes
14automated checks in CI
2architecture decision records
1 cmdto start Kafka locally
04 / EXPERIENCE
A career shaped by consequential systems.
AUG 2024 — PRESENT
Software Development Senior Analyst · Java
Building and maintaining Java backend functionality for pension benefit calculation workflows in financial services.
- Translate plan provisions and complex business rules into reliable, testable calculation logic.
- Investigate edge cases and collaborate across analysis, QA, and delivery to protect correctness in high-impact financial outcomes.
- Improve maintainability through automated tests, code review, and clear technical reasoning.
JUN — DEC 2023
API Marketplace Developer
Worked on the platform and delivery systems behind an enterprise API marketplace.
- Built GitHub Actions and GCP deployment workflows, managed secrets, and integrated Maven into CI/CD.
- Developed API token-generation functionality and strengthened regression coverage with Serenity BDD.
- Used Splunk to investigate anomalies across deployed services.
DEC 2019 — JUN 2022
Senior Systems Engineer
Built Java and Spring services for India's national GST e-invoicing ecosystem, supporting high-volume invoice registration and regulatory reporting.
- Implemented PAN-based Annual Aggregate Turnover (AATO) eligibility rules used to enable taxpayers for e-invoicing requirements.
- Built Kafka-backed event workflows and REST integrations connecting taxpayer and GSP services with the NIC Invoice Registration Portal, GST, and e-Way Bill systems.
- Strengthened secure access, migrations, persistence, and automated coverage with Spring, Hibernate, SQL, JUnit, Mockito, Testcontainers, and JMeter.
05 / OPEN SOURCE
Code that shipped upstream.
Contributions to tools used across the Java ecosystem—reviewed by the maintainers who know them best.
Spring Boot
RELEASE CONTRIBUTOR
Spring Boot 4.0
Credited among the community contributors who helped deliver the Spring Boot 4.0 release.
View the sourceJenkins
MERGED · JAVA + JUNIT
Clearer queue diagnostics
Improved queue diagnostics by truncating verbose blockage messages, backed by boundary and stress tests.
View the sourceApache Maven
MERGED · BUILD TOOLING
Predictable test logging
Removed conflicting SLF4J test providers and standardized tests on Maven’s own logging provider.
View the source06 / CONTINUOUS LEARNING
Technical depth, broadened by data.
EDUCATION
Business Intelligence Analytics
Nova Scotia Community College · 2024
Full Stack Application Development
Nova Scotia Community College · 2023
Bachelor of Technology, Engineering
APJ Abdul Kalam Technological University · 2019
SELECTED CREDENTIALS
GitHub Actions Certified
GitHub · Valid through 2027
REST API · Intermediate
HackerRank · 2024
SQL · Advanced
HackerRank · 2024
07 / GET IN TOUCH
Let’s build software people can depend on.
I’m interested in Java backend roles where domain depth, engineering discipline, and thoughtful collaboration matter.
steve@stevearmstrong.org