OpenSearch ingestion pipeline rebuild
Indexing 1M records went from 240 minutes to 6 minutes (~40× faster), and large runs stopped starving new indexing.
240m → 6m~40× fasterNo backfill stalls
Lisbon, Portugal · open to select work
Backend · Cloud · Systems
I build backend systems, cloud infrastructure, and software products that hold up at real-world scale.
Senior software engineer focused on high-scale backend services, AWS architecture, and data infrastructure — and building software products around B2B systems and automation.
ingestion-pipeline.cdc240m → 6m
// replaced DMS + Kinesis · decoupled backfills from live writes · ~40× faster
Node.js/TypeScript/AWS/AWS CDK/Lambda/SQS/Kafka / MSK/Debezium/OpenSearch/MySQL / Aurora/PostgreSQL/Sequelize/Redis/Docker/Serverless/Distributed systems/Node.js/TypeScript/AWS/AWS CDK/Lambda/SQS/Kafka / MSK/Debezium/OpenSearch/MySQL / Aurora/PostgreSQL/Sequelize/Redis/Docker/Serverless/Distributed systems/
- $0M+
- MRR milestone reached by multiple SaaS products I helped scale
- 0×
- faster search ingestion after pipeline rebuild
- 0B+
- rows in a production database table I optimized for reporting
- ~$0k
- annual cloud spend saved after CDK migration and right-sizing
Work / Expertise
What I'm good at, with the specifics.
01
Backend Engineering
Node.js / TypeScript services that carry real business logic and real traffic. Migrated a core message-routing service (the 'Sorter') from PHP to Node.js, improving performance and maintainability of the most critical component in the system.
02
Cloud Architecture
AWS-native, event-driven systems built as code. Owned an end-to-end migration of high-scale infrastructure from console-managed resources to AWS CDK — compute, messaging, data, ingress, and network — with reusable patterns and CI/CD guardrails that make changes reviewable and drift-resistant.
03
Data Pipelines
Change-data-capture and streaming pipelines that keep systems in sync. Built a Debezium → MSK (Kafka) → OpenSearch Ingestion path that replaced a slower DMS + Kinesis design and removed the bottleneck where large runs blocked new indexing.
04
Search Infrastructure
OpenSearch ingestion and indexing at scale. Re-architected the ingestion pipeline so indexing 1M records dropped from 240 minutes to 6 minutes — roughly 40× faster — while keeping new writes flowing during large backfills.
05
Database Performance
MySQL / Aurora and PostgreSQL under genuine pressure. Optimized a production reporting table with over a billion rows using denormalization, partitioning, cursor pagination, replicas, caching, indexing, and query guardrails.
06
Reliability & Systems Design
Distributed, event-driven systems designed to be observable and affordable. Rebuilt a customer-facing Inbox to cut response times by ~95%, and reduced monthly cloud spend from $30k to $22.5k through IaC-led infrastructure modernization, right-sizing, and architecture changes.
Selected system work
A few systems, described by what changed.
Console-to-CDK infrastructure migration
Infrastructure changes became repeatable and code-reviewed, while monthly spend dropped from $30k to $22.5k (~$90k/year) after CDK migration, usage-based container right-sizing, Graviton moves, and opt-in AWS cost controls.
$30k → $22.5k / mo~$90k / yr savedRepeatable IaC
Building something that needs to scale?
I take on senior backend and cloud work, and selected product builds. Tell me what you're building.