Skip to content
View danielcruzdev's full-sized avatar
🎯
Focusing
🎯
Focusing

Block or report danielcruzdev

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Content in all repositories owned by your account will be closed.
Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
danielcruzdev/README.md

Daniel Cruz

Software Architect · .NET · Distributed Systems

Building financial systems that stay correct when things go wrong.

LinkedIn LeetCode


About

Software Architect and Developer at Banco Sofisa, working on backend systems in the .NET ecosystem.

My focus is the part of software that only shows up under failure: consistency across service boundaries, idempotency, compensation, and the observability you need to prove any of it works. Most of my repositories here are built around a single question — what happens when a dependency goes down mid-transaction?

  • 🏗️ Distributed systems, event-driven architecture, and pragmatic API design
  • 📐 I document decisions as ADRs with the alternatives I rejected and why
  • 📊 I state non-functional requirements as numbers and test against them
  • 🌍 Based in Brazil (UTC−3) · Working in English and Portuguese

Featured Work

Two-service cash flow platform built around one hard requirement: the transaction service stays available even when the consolidation service is completely down — and there is a test that proves it.

Transactional Outbox, idempotent consumers, compensating reversals, circuit breakers, and liveness/readiness separation. Ships with 9 ADRs, C4 diagrams, failure-path sequence diagrams, an operational runbook, and k6 load tests with SLO thresholds enforced in CI.

Measured Result SLO
p95 latency 1.98 ms < 100 ms
Error rate @ 50 req/s 0% ≤ 5%
Read-model consistency lag ~2.2 s < 5 s

.NET 10 CQRS Outbox RabbitMQ PostgreSQL Redis Polly Testcontainers k6


Double-entry bookkeeping API where balances can never drift: entries are append-only, corrections happen through reversals rather than updates, and concurrent writes are resolved with optimistic concurrency. Integration tests run against a real PostgreSQL instance.

.NET 10 Minimal APIs PostgreSQL EF Core Vertical Slice Testcontainers


An AI agent and Model Context Protocol server written from scratch — no SDK. JSON-RPC 2.0 over stdio with Content-Length framing, tool-calling orchestration, path-traversal protection on file access, and 18 scenario tests covering realistic agent workflows.

.NET 10 MCP JSON-RPC 2.0 AI Agents OpenAI API xUnit


Messaging abstraction that lets application code publish and subscribe without binding to a provider SDK. Factory-based multi-broker support, pluggable serialization, and a correlation-aware message envelope. Published to NuGet.

.NET 10 Kafka RabbitMQ Azure Service Bus NuGet


Task Manager API where each use case owns its endpoint, handler, validation, and tests — organized by feature instead of by technical layer.

.NET 10 Vertical Slice CQRS MediatR Carter FluentValidation


Working comparison of the four ASP.NET Core rate limiting strategies — Fixed Window, Sliding Window, Token Bucket, and Concurrency Limiter — with the trade-offs of each.

.NET ASP.NET Core Minimal APIs Rate Limiting


Tech

Core — C# · .NET 10 · ASP.NET Core · Minimal APIs · EF Core · Dapper

Architecture — CQRS · Vertical Slice · DDD · Transactional Outbox · Saga & compensation · Idempotency

Data & Messaging — PostgreSQL · Redis · RabbitMQ · Kafka · Azure Service Bus

Reliability & Ops — OpenTelemetry · Polly · Docker · GitHub Actions · k6 · Testcontainers · xUnit

AI — Model Context Protocol · OpenAI API · tool-calling agents


Currently

Working through the operational side of the cash flow platform — Kubernetes deployment, autoscaling driven by queue depth, and infrastructure as code.


Contact

Open to conversations about distributed systems, architecture, and interesting problems.

LinkedIn


contribution grid snake animation

Pinned Loading

  1. dotnet-cqrs-cashflow dotnet-cqrs-cashflow Public

    Controle de fluxo de caixa em .NET 10 com CQRS, Transactional Outbox e RabbitMQ. O serviço de lançamentos continua aceitando escritas com o consolidado fora do ar

    C#

  2. dotnet-mcp-server dotnet-mcp-server Public

    AI Agent + MCP Server built from scratch in .NET 10 — JSON-RPC 2.0 over stdio, tool-calling loop with OpenAI, and a clean layered architecture. Portfolio & study project.

    C#

  3. Lib.MeshBus Lib.MeshBus Public

    🚌 Unified .NET 10 messaging abstraction library. Publish & subscribe to Kafka, RabbitMQ, Azure Service Bus through a single interface. Swap brokers with one line change — zero impact on business co…

    C#

  4. dotnet-api-fintech-ledger dotnet-api-fintech-ledger Public

    A double-entry fintech ledger API built with ASP.NET Core .NET 10 and PostgreSQL, featuring idempotent transactions, append-only reversals, optimistic concurrency, and integration tests.

    C#

  5. dotnet-api-rate-limit dotnet-api-rate-limit Public

    Exemplo prático de implementação de Rate Limiting em .NET 9 utilizando Minimal API e o middleware nativo do ASP.NET Core, com explicações claras dos principais algoritmos.

    C#

  6. dotnet-vertical-slice-architecture dotnet-vertical-slice-architecture Public

    A .NET 10 REST API showcasing Vertical Slice Architecture with MediatR, Carter, FluentValidation, and EF Core (PostgreSQL). Features organized by use case, not by layer — each slice owns its endpoi…

    C#