Skip to main content

Loopai v0.3 Released: Multi-Language SDK Ecosystem

· 3 min read
Caveman
Caveman
Lead Developer, Loopai Framework

We're thrilled to announce Loopai v0.3, featuring a complete multi-language SDK ecosystem with production-ready client libraries for .NET, Python, and TypeScript.

🎉 Release Highlights​

Three Production-Ready SDKs​

.NET Client SDK​

  • Package: Loopai.Client (NuGet)
  • Features: ASP.NET Core DI, Polly v8 retry, comprehensive logging
  • Test Coverage: 92% lines, 88% branches
// Dependency Injection
builder.Services.AddLoopaiClient(options =>
{
    options.BaseUrl = "http://localhost:8080";
    options.Timeout = TimeSpan.FromSeconds(60);
});

// Usage
var result = await _loopai.ExecuteAsync(taskId, inputData);

Python SDK​

  • Package: loopai (PyPI)
  • Features: Full async/await, Pydantic v2 models, context manager
  • Test Coverage: 94% lines, 90% branches
async with LoopaiClient("http://localhost:8080") as client:
    result = await client.execute(
        task_id=task_id,
        input_data={"text": "Sample input"}
    )

TypeScript SDK​

  • Package: @loopai/sdk (npm)
  • Features: Full type safety, promise-based async, browser support
  • Test Coverage: 93% lines, 89% branches
const client = new LoopaiClient({
  baseUrl: 'http://localhost:8080',
});

const result = await client.execute({
  taskId: taskId,
  input: { text: 'Sample input' },
});

📊 Quality Metrics​

Integration Testing​

We've implemented comprehensive integration testing:

  • 42 Integration Tests: 14 tests per SDK
  • 100% Pass Rate: All tests passing
  • Cross-SDK Compatibility: Verified interoperability
  • CI/CD Ready: GitHub Actions workflows
SDKTestsPassSuccess RateAvg Response Time
.NET1414100%45.2ms
Python1414100%43.8ms
TypeScript1414100%44.5ms
Total4242100%44.5ms

Performance Benchmarks​

All SDKs deliver consistent, high performance:

  • Average Response Time: <45ms
  • Throughput: ~34 requests/second per SDK
  • Memory Usage: 52-62 MB average
  • Execution Latency: <1ms (p99)

🚀 New Features​

Batch Operations​

All SDKs support efficient batch processing:

# Python example
results = await client.batch_execute(
    task_id=task_id,
    items=[
        {"id": "1", "input": {"text": "First"}},
        {"id": "2", "input": {"text": "Second"}},
    ],
    max_concurrency=10
)

Automatic Retry​

Built-in retry logic with exponential backoff:

  • Configurable max retries
  • Smart backoff strategy
  • Network failure resilience

Type Safety​

Full type coverage across all SDKs:

  • .NET: Strong typing with generics
  • Python: Pydantic v2 models with type hints
  • TypeScript: Complete TypeScript definitions

📚 Documentation​

Complete documentation for all SDKs:

Each SDK includes:

  • Installation instructions
  • Quick start guide
  • API reference
  • Code examples
  • Best practices

🛠️ Breaking Changes​

None! This is a new SDK release with no breaking changes to existing APIs.

📦 Installation​

.NET​

dotnet add package Loopai.Client

Python​

pip install loopai

TypeScript​

npm install @loopai/sdk

🔜 What's Next: v0.4​

We're now focusing on infrastructure enhancements:

  1. Streaming API

    • Real-time execution results
    • Server-Sent Events (SSE)
    • Available in all SDKs

  2. Go SDK

    • Native Go client
    • Context support
    • Goroutine-safe operations

  3. Performance Optimization

    • Response caching layer
    • Connection pooling
    • Reduced memory footprint

🙏 Acknowledgments​

Special thanks to:

  • All contributors who tested the SDKs
  • The community for valuable feedback
  • Early adopters who reported issues

📝 Changelog​

For a complete list of changes, see:

💬 Feedback​

We'd love to hear your thoughts!

Download Loopai v0.3 today and start building with production-ready SDKs!