Articles by Aarav Patel

Learn how to design cost-aware multi-model routing for enterprise AI APIs using model quality, latency, token costs, availability, fallback policies, and workload-specific routing strategies.

Compare agentic document extraction with traditional OCR pipelines by measuring extraction accuracy, processing latency, token costs, error rates, and human review requirements across enterprise documents.

Build production-ready document processing pipelines with Azure Content Understanding using semantic chunking, agentic document understanding, structured extraction, synchronous APIs, and modern GPT models.

Benchmark AI model routing with Microsoft.Extensions.AI by comparing model selection strategies across latency, cost, token usage, response quality, throughput, and routing overhead in .NET applications.

Learn how to build automated NuGet vulnerability gates for .NET CI pipelines that detect vulnerable dependencies and enforce security policies before deployment.

Benchmark .NET HTTP connection eviction under long-lived traffic to measure stale connections, DNS changes, failover behavior, latency, and connection lifetime in production workloads.

Explore password-protected ZIP archives in .NET and learn how to securely package, transfer, and protect application data using the latest ZIP security capabilities.

Test the new .NET 11 DNS APIs under realistic failover conditions, measuring DNS resolution latency, caching behavior, recovery time, and service-discovery reliability.

Learn how .NET 11 HTTP connection eviction helps prevent stale connections and improve reliability, latency, and connection-pool behavior in production applications.

Explore HTTP request compression in .NET 11 through practical benchmarks measuring bandwidth usage, CPU overhead, latency, and throughput across different compression scenarios.

Prevent AI agent failures by testing MCP server schema drift and tool compatibility. Ensure robust contracts for reliable AI interactions.

Unlock enterprise AI with Agent Skill Registries. Discover, govern, and version reusable skills for .NET platforms, ensuring robust lifecycle management.

Benchmark MCP C# SDK 2.0 for .NET AI apps. Measure latency, throughput, concurrency, and resource usage under realistic tool workloads.

Master declarative AI workflow testing with failure injection. Ensure robust error handling, recovery, and reliability in production.

Securely design multi-tenant AI agent skills with tenant isolation, authorization, filtering, and audit logging for enterprise platforms.

Benchmark stateless MCP servers behind load balancers to measure throughput, latency, resource consumption, and scaling behavior for optimal .NET deployments.

Migrate MCP C# servers from SDK 1.x to 2.0 with this guide. Learn about stateless HTTP, protocol negotiation, and API changes for a smooth transition.

.NET 11's Runtime Async revolutionizes async programming with runtime-managed suspension, boosting performance and debugging.

Unlock MCP HTTP scalability with C# SDK 2.0. Embrace stateless design for easier horizontal scaling and leverage familiar ASP.NET Core infrastructure.

Benchmark C# 15 union types in real apps. Explore performance, allocations, boxing, and compare with interfaces for production readiness.

Secure AI coding agents with capability-based access. Enforce least privilege, scope permissions, and audit actions for robust AI development security.

Build robust AI-powered APIs in ASP.NET Core with structured output validation. Ensure reliable data integration and prevent downstream errors.