Arika

Thoughtful Writing on Architecture, AI & Engineering

Read the Blog Browse Series

Series

Structured deep-dives, read from start to finish

View all →

SERIES18 parts · ~114 min

Java Release Features

Explore lambda expressions, streams, and functional interfaces that transformed Java development.

Start reading →·View all 18 parts

SERIES17 parts · ~146 min

Legacy Modernization

How to define modernization objectives, model ROI, manage risk, and secure executive sponsorship with a pragmatic AI-assisted playbook.

Start reading →·View all 17 parts

Designing for Observability from Day One

Observability is not something you bolt on after an incident. This is the practical guide to treating logs, metrics, and traces as first-class design concerns — with the instrumentation patterns that make systems debuggable in production.

Apr 30, 2026·9 min read

EngineeringObservabilityMonitoring

Engineering

Technical Debt You Should Actually Pay Down (and What to Leave)

Not all technical debt is worth addressing. A practical framework for classifying debt by urgency and delivery impact — and the honest case for leaving most of it alone.

Apr 28, 2026·8 min read

EngineeringTechnical DebtArchitecture

Java

Java Performance Tuning Without a PhD

A practical guide to JVM performance tuning — profiling methodology, GC algorithm selection, heap sizing, JIT compilation hints, and the flags that actually matter in production.

Apr 25, 2026·8 min read

JavaPerformanceJVM

Java

Virtual Threads vs Reactive: When to Pick Each

Java 21's virtual threads and Project Reactor both solve the I/O concurrency problem. Here is the practical guide to choosing between them — performance trade-offs, migration paths, and where each genuinely excels.

Apr 22, 2026·7 min read

JavaVirtual ThreadsReactive

AI-Assisted Code Review: What Works, What Doesn't

An honest assessment of AI code review tools after two years in production — where they catch real bugs, where they produce noise, and how to integrate them without destroying reviewer trust.

Apr 20, 2026·8 min read

AICode ReviewEngineering

Prompt Engineering for Developers: Patterns That Actually Work

Skip the theory. Here are the concrete prompt patterns — few-shot, chain of thought, structured output, role prompting, self-critique — that make LLMs reliable in production software.

Apr 17, 2026·9 min read

AIPrompt EngineeringLLM