Programming News and Topics: Latest News on Computer Programming from Techminis

A naukri.com initiative

New

Home

Programming News

Microsoft

179

Image Credit: Microsoft

Evaluating content safety in your .NET AI applications

The Microsoft.Extensions.AI.Evaluation.Safety package has been added to the Microsoft.Extensions.AI.Evaluation libraries for detecting harmful content in AI-generated applications.
The safety evaluators, powered by Azure AI Foundry Evaluation service, can be seamlessly integrated into existing workflows.
Steps to set up Azure AI Foundry for safety evaluations include creating an Azure subscription, resource group, AI hub, and project.
A C# example shows how to configure and run safety evaluators to check AI responses for various criteria.
Running unit tests and generating reports can be done using Visual Studio, Visual Studio Code, or the .NET CLI.
The API usage samples cover scenarios like evaluating content safety of AI responses with images and running safety and quality evaluators together.
Updates to Microsoft.Extensions.AI.Evaluation libraries include enhanced quality evaluators and improved reporting functionality.
New reporting features allow searching and filtering scenarios using tags, viewing rich metadata for metrics, and tracking historical trends.
Developers are encouraged to explore the content safety evaluators and provide feedback for further enhancements.
The aim is to elevate the quality and safety of AI applications using these tools.
The post emphasizes continuous improvement and invites engagement for better utilization of the libraries.

Read Full Article

10 Likes

Logrocket

381

Image Credit: Logrocket

Profit center vs. cost center: How company structure affects engineering

Understanding the differences between profit and cost centers in engineering teams is crucial for developers, as each operates with distinct mentalities and structures.
In profit centers, like engineering teams supporting revenue-generating products, investment is seen as beneficial for innovation and revenue growth.
Profit centers own specific product scopes, with squads led by engineering managers and clear ownership structures to drive continuous improvement.
However, drawbacks of profit centers include potential team silos and the need for engineers to work on cross-functional projects for career advancement.
On the other hand, in cost center models, engineers are viewed as resources for projects that support the company's main revenue-generating departments.
Projects in cost centers may follow a more waterfall methodology, with developers moving to different projects once their current assignment is complete.
Pros of cost centers include working on diverse projects, slower pace for better work-life balance, modern tech stacks, and greater job security due to risk aversion.
However, the downsides of cost centers include lower pay, less exciting work, lack of ownership sense, and a slower process for addressing technical debt.
Choosing between profit and cost centers depends on individual preferences, with stability being a key factor that influences career choices.
Understanding how engineering teams are positioned within an organization can help developers, whether early in their career or with long-term goals, make informed career decisions.

Read Full Article

22 Likes

Pymnts

320

Image Credit: Pymnts

Microsoft to Retire Bing Search APIs, Promote Azure AI Agents

Microsoft will retire Bing Search APIs on Aug. 11, as announced on Monday.
Customers are advised to consider using Grounding with Bing Search as an alternative, integrated with Azure AI Agents.
Grounding with Bing Search allows Azure AI Agents to incorporate real-time public web data in responses, focusing on chatbots.
The shift to AI-powered system is to meet market demand; Microsoft offers support for affected customers.

Read Full Article

18 Likes

Medium

322

Image Credit: Medium

Integrate Sign-In with Apple via Firebase in a Compose Multiplatform project

Integrating Sign-In with Apple in a Compose Multiplatform project using Firebase wrapper made for KMP: GitLive.
Implemented Google Sign-In for Android and iOS, but faced challenges with Apple Sign-In without SwiftUI.
Utilized Clean Architecture principles, creating interfaces and wrappers for different social media providers like Apple.
Incorporated native Swift code into Kotlin Multiplatform project using spm4Kmp plugin for deep integration with iOS frameworks.

Read Full Article

19 Likes

Dev

190

Image Credit: Dev

A Professional Looking API

The OpenAPI spec is crucial for defining how consumers interact with an API and can enhance developer experience.
Version 1 and Version 2 of an endpoint are visualized to demonstrate improvements in documentation and client integration.
API versioning using URL-based approach and Swagger enhances clarity and long-term maintainability.
Refactored endpoints with task-based actions like SetShippingInfo result in cleaner docs, easier client integration, and a more professional API impression.

Read Full Article

11 Likes

Dev

367

Image Credit: Dev

One-Click Deployment of n8n on Clawcloud Run: Build Enterprise-Grade Automation Workflows with Zero Code

n8n and Clawcloud Run offer an efficient solution for workflow automation with unique features and advantages.
Clawcloud Run provides zero-cost entry, high-performance configuration, and simplified operations for deploying applications.
n8n offers open-source control, flexible extensibility, and a hybrid execution model for diverse automation needs.
By following a step-by-step deployment process, users can set up n8n on Clawcloud Run and optimize it for advanced functionalities like domain binding, security hardening, and third-party service integrations.

Read Full Article

22 Likes

Dev

127

Image Credit: Dev

LocalLLMClient: A Swift Package for Local LLMs Using llama.cpp and MLX

LocalLLMClient is a Swift package that enables the use of local Large Language Models (LLMs) on Apple platforms.
It supports models like MobileVLM-3B (llama.cpp) and Qwen2.5 VL 3B (MLX) for iOS and macOS.
Features of LocalLLMClient include GGUF / MLX model support, iOS and macOS compatibility, streaming API, and experimental multimodal support.
The library leverages llama.cpp and Apple MLX backends with a unified interface, allowing interactions with LLMs.
Basic usage involves importing the necessary components, downloading models from Hugging Face, and generating text using specified parameters.
Text generation examples demonstrate creating stories with a cat protagonist and generating song lyrics from image inputs.
Additional features in LocalLLMClientUtility like FileDownloader offer functionalities for managing model downloads efficiently.
The project emphasizes on-device processing for privacy and cost-effectiveness, thanking the community for supporting local LLM development.
Overall, LocalLLMClient aims to provide a convenient option for Swift developers wanting to experiment with AI on Apple platforms.
Supporting the project through contributions or stars on GitHub is encouraged.

Read Full Article

7 Likes

Dev

154

Image Credit: Dev

How to Loop Multiple WAV Stems with AudioKit in Swift?

Playing multiple audio tracks simultaneously in your Swift app using AudioKit's AudioPlayer can create immersive audio experiences.
AudioKit's AudioPlayer class is powerful for playback but might have issues with looping WAV stems due to initialization, timing conflicts, or start and stop calls.
To loop WAV stems with AudioKit, you need to ensure correct initialization of audio players, start them properly, manage playback states, and check AudioKit setup.
Additionally, you can apply audio effects to players, troubleshoot stopping after the first loop, and consider performance implications of multiple players.

Read Full Article

9 Likes

Dev

149

Image Credit: Dev

Spring Boot Integration Testing

Spring Boot Integration Testing focuses on testing real applications in a cohesive manner using Spring's tools.
While JUnit and Mockito are great for isolated testing, Spring's integration testing framework tests the whole application.
Spring MockMvc allows testing controllers without server deployment, simulating the full Spring MVC lifecycle.
@WebMvcTest is for testing the web layer, focusing on MVC components and configuring MockMvc.
@SpringBootTest creates a full application context for testing interaction between all beans and providing a realistic environment.
@Import allows supplementing test slice with additional beans, while @ContextConfiguration gives explicit control over context setup.
Common issues in integration testing include package structure for main configuration and deprecated MockBean usage.
Mocking all repositories used by the Service class can prevent UnsatisfiedDependencyException during context initialization.
Integration tests example includes testing a JobController that interacts with a JobService calling a Mongo Repo.
Integration tests cover scenarios like successful job retrieval, empty results, missing query parameters, and service exceptions.

Read Full Article

9 Likes

Javacodegeeks

Image Credit: Javacodegeeks

HATEOAS + Spring Security: Why Some Links Are Missing for Certain Roles

HATEOAS (Hypermedia As The Engine Of Application State) combined with Spring Security allows for dynamically shaping API responses based on user roles.
Spring HATEOAS enriches REST API responses with navigable links while Spring Security provides authentication and authorization.
Common reasons for missing HATEOAS links include manual role-based filtering, missing or incorrect role checks, and static link definitions.
A better approach involves role-aware link building using Spring HATEOAS and Spring Security together, ensuring links adapt based on user roles.

Read Full Article

2 Likes

Dev

Image Credit: Dev

Grok 3 API - Reasoning Tokens are Counted Differently

OpenAI includes reasoning tokens in usage.completion_tokens
xAI doesn't include reasoning tokens in usage.completion_tokens
To get the total tokens with OpenAI or Deepseek R1, use the completion_tokens field. With xAI, sum the two values to get correct totals.
Grok API provides a compatible OpenAI endpoint but xAI deviates from OpenAI's reasoning tokens accounting approach.

Read Full Article

Dev

425

Image Credit: Dev

React Native 0.79+ New Architecture: Bridging with Android Native Modules

React Native allows developers to interact with platform-specific code using Native Modules for accessing native Android features.
The blog post details creating a native module in Android and bridging it with React Native, including calling native methods and sending data between React Native and Android.
Steps provided involve defining interface in Kotlin class, registering the native module with React Native using a package, and calling native methods from React Native in the app.
By following these steps, developers can effectively bridge JavaScript with the native Android layer to integrate SDKs or platform-specific features not readily available in React Native.

Read Full Article

25 Likes

Logrocket

291

Image Credit: Logrocket

How UX designers can influence AI behavior (and why they must)

Generative AI is transforming UX design by shifting focus from minor details to holistic system behavior.
Understanding concepts like statistical mechanics enables designers to create more effective AI experiences.
Designers must shift their focus from micro to macro levels when designing with generative AI.
It is essential for designers to influence AI behavior to ensure beneficial outcomes and user-centered design.
AI functionality in digital tools requires designers to engage with the mechanics of AI for responsible operation.
UX designers play a crucial role in shaping AI inputs, outputs, and behavior through human-centered design.
Designers need to focus on transparency, ethics, and user advocacy when designing AI systems.
From interface design to architecting AI behavior, designers need to adapt to new responsibilities and deliverables.
The shift towards influencing AI behavior requires a change in methods of documentation and communication for UX designers.
Collaboration with various stakeholders is key for designing AI that benefits users and avoids harm.

Read Full Article

17 Likes

Medium

247

Image Credit: Medium

Java’s Concealed Weaknesses: Insights from a Seasoned Coder

Java, a popular and widely-used programming language, has hidden drawbacks that may surprise developers.
Despite its power and reliability, Java has some weaknesses that can slow down developers, as highlighted by a seasoned coder.
While Java taught object-oriented programming and provided a structured foundation, its limitations became apparent as the coder explored other languages.
This blog offers honest insights into the concealed weaknesses of Java, providing valuable perspectives for both experienced and upcoming Java developers.

Read Full Article

14 Likes

The Pragmatic Engineer

3.2k

Image Credit: The Pragmatic Engineer

The Pulse #134: Stack overflow is almost dead

Stack Overflow is facing a significant decline in the number of questions asked per month, reaching levels last seen when it was launched in 2009.
Google's CEO recently involved in customer support, AI dev tools company valuations are soaring, and OpenAI provides more clarity on its nonprofit status.
Microsoft recently laid off 6,000 employees, with a potential indication of more regular job cuts in the future.
Google has implemented a new bonus system where high performers receive more bonuses, while low performers receive less, in line with other tech companies' practices.

Read Full Article

22 Likes

For uninterrupted reading, download the app

Programming News

Discover more