Consumer Lag Is the Wrong Finish Line
A recovery verdict based on terminal outcomes, idempotency, and reconciliation of business state.
Read on DZoneWriting & publications
Engineering, written down.
Technical articles on event-driven recovery, Java and Kafka, and the controls that make AI useful in engineering.
Published work
Technical articles
The Best AI Fix Isn't Adding More: It's Taking Away
Reducing tool and context overload in agent workflows. Published with a co-author; both bylines appear on the article.
Read on HackerNoonTesting Kafka Consumer Recovery Contracts With Testcontainers and JUnit
Contract tests for duplicate delivery, malformed messages, and the boundary between side effects and offset commits.
Read on FoojayAI Incident Response Needs A Control Plane, Not a Chatbot
Operational controls for incident-response agents: evidence, authority, and approval before action.
Read on HackerNoonDesigning Replay-Safe CDC Pipelines With Kafka, Debezium, and Recovery Contracts
Recovery contracts connecting CDC, projections, bounded replay, and reconciliation.
Read on DZoneAI Agents Should Bring Receipts, Not Just Answers
Evidence trails and approval boundaries for agents working inside engineering systems.
Read on HackerNoonBuilding Idempotent Java Consumers for Replay-Safe Kafka Pipelines
Durable acceptance records, stable identities, and duplicate suppression for Java consumers.
Read on Java Code GeeksBuilding Event-Driven Systems That Can Recover With Confidence
How authoritative history, ordering boundaries, and controlled replay make recovery testable.
Read on HackerNoonEditorial recognition
Interviews, features, and distribution
Editorial coverage is listed separately from authored articles.
Writers Spotlight
An editorial interview about recoverable systems, AI guardrails, and engineering practice.
Read the Writers Spotlight interviewHomepage features
DZone confirmed homepage placement for the CDC article on September 2 and Consumer Lag Is the Wrong Finish Line on October 13.
Read Consumer Lag Is the Wrong Finish LineTech Brief Podcast
HackerNoon notified me that Building Event-Driven Systems That Can Recover With Confidence was distributed through its Tech Brief Podcast.
Read the associated articleDaily TechBeat ranking
HackerNoon notified me that Building Event-Driven Systems That Can Recover With Confidence ranked among its most-read stories of the preceding week and appeared in its Daily TechBeat.
Read the associated articleIndustry articles
Additional bylined writing
Reliability Classes for Event-Driven Platforms: Lessons from Billing, Retail, and Security Systems at Scale
A framework for discussing reliability requirements across different operational domains.
Read on DataconomyDeterministic Coverage-First Allocation: A Transparent Retail Approach
An approach to retail allocation centered on explicit constraints, reproducibility, and planner control.
Read on Coruzant Technologies