Niharika Gupta's contributions

Learn how to build automated GitHub Actions quality gates using code quality checks to detect issues and enforce consistent software quality in CI/CD pipelines.

Benchmark AI agent tool selection with 10, 100, and 500 available tools to measure selection accuracy, latency, token usage, and wrong-tool calls.

Compare .NET 10 LTS and .NET 8 runtime performance using practical benchmarks for CPU usage, memory consumption, throughput, latency, and application workloads.

Learn how to use CodeQL to test GitHub Actions workflows for injection vulnerabilities and strengthen CI/CD security against malicious workflow inputs.

Learn how GitHub Copilot usage metrics can measure AI agent adoption across development teams, identify usage patterns, and improve enterprise AI strategy.

Learn how NuGet Trusted Publishing enables keyless .NET package releases using modern identity-based authentication to reduce long-lived credential and supply chain risks.

Learn how to automate NuGet API key rotation in GitHub Actions and safely manage 30-day credentials without disrupting .NET package publishing workflows.

Benchmark Microsoft Foundry Hosted Agents by measuring sandbox startup time, isolation overhead, compute usage, memory, filesystem performance, and agent execution efficiency.

Learn how Microsoft Foundry Toolboxes can centralize AI tool governance through authentication, lifecycle management, versioning, permissions, discovery, and enterprise controls.

Benchmark end-to-end voice agent latency from speech recognition through model response and tool execution to speech synthesis, identifying bottlenecks across the complete conversational AI pipeline.

Build real-time voice agents with .NET and Azure Voice Live for low-latency conversational experiences using speech processing, AI models, tools, memory, knowledge, and agent workflows.

Design resilient AI fallback chains for provider outages, model failures, throttling, timeouts, latency spikes, and quality degradation while maintaining reliable .NET application behavior.

Benchmark RoutingChatClient against static model selection in .NET by measuring routing overhead, latency, token usage, response quality, cost, and performance across different AI workloads.

Learn how to design secure identity boundaries for AI agent tool calls using user identity, delegated permissions, authorization policies, and least-privilege access.

Compare HTTP compression algorithms in ASP.NET Core APIs using real-world benchmarks for response size, CPU consumption, latency, throughput, and network efficiency.