Node.js (TypeScript), PHP • Microservices • Distributed Systems • Redis • Elasticsearch • AWS • Azure • Docker/Kubernetes Open to full-time roles: remote, hybrid, or on-site with relocation support in select countries • Timezone: UTC-3 • rasantos.dev

• Build and scale high-traffic microservices and APIs.
• Event-driven pipelines (AWS SQS/SNS), caching (Redis), and search (Elasticsearch).
• Applied AI integration with LangChain and OpenAI API for RAG and automation, not model training.
• Performance engineering: latency, throughput, error rate, and profiling.
• Cloud and DevOps: AWS, Azure AKS, Docker, CI/CD, observability (Prometheus, Grafana).

• Average latency: about 600 ms to 194 ms (68% lower)
• p99 latency: 3.6 s to 446 ms (88% lower)
• 5xx errors: 4% to 0.1% (97.5% lower)
• LLM API cost: up to 80% lower (caching + prompt strategy + RAG)
• Auth overhead: 95% lower with token caching

Refactored a mission-critical freight domain into a clean, isolated service.
Results: 68% lower latency, 97.5% fewer errors, 40% higher throughput.
Stack: Node.js (TS), Redis, Elasticsearch, MySQL, Docker/Kubernetes, Prometheus/Grafana.

Improved logs and traceability and enabled minute-level quotation granularity.
Focus: reliability, observability, and data correctness across services.

High-performance REST API with PostgreSQL/MySQL and Redis caching.
Result: about 95% cache hit rate and sub-second responses.
Repo/Docs: on request (public endpoints and rate limits documented).

RAG pipeline using LangChain and OpenAI for petition generation.
Results: 10x faster document prep, 80% lower inference cost.
Focus: prompt strategy, retrieval quality, and cost controls.

Event-driven architecture with BullMQ, Redis, and Auth0.
Results: 95% lower auth overhead and horizontally scalable design.

Applied NLP for complaint categorization and sentiment.
Results: 90% less manual analysis time and 10k+ documents processed.

Languages and Runtime: Node.js (TypeScript/JavaScript), PHP (Laravel)
Messaging and Data: AWS SQS/SNS, Redis, Elasticsearch, MySQL/PostgreSQL
Applied AI: LangChain, OpenAI API, RAG, prompt engineering, cost optimization
Cloud and Ops: AWS, Azure AKS, Docker, CI/CD, FluxCD, Prometheus, Grafana
Testing and Quality: Jest, k6, OpenAPI/AsyncAPI, tracing and logging by default

• Measure before optimizing - data-driven performance work.
• Design for failure - retries, backpressure, idempotency.
• Build for observability - logs, metrics, tracing by default.
• Optimize for ROI - performance and efficiency in balance.

• rasantos.dev
• linkedin.com/in/rasantos93
• x.com/rasantosdev93
• rodrigo@rasantos.dev

Note: All projects are anonymized case studies. No proprietary code or confidential data is included.