Google announced on Wednesday the discovery of a new strain of malware, dubbed “LOSTKEYS”, which has been attributed to the Cold River hacking group—an actor previously linked to Russian state interests.

The revelation was detailed in a blog post by Google’s Threat Intelligence Group (GTIG), highlighting the evolving tactics of a threat actor known for targeting high-profile geopolitical entities.

A New Chapter in Cold River’s Toolkit

According to Wesley Shields, a security researcher at Google GTIG, LOSTKEYS represents “a new development in the toolset” of Cold River. The malware is capable of exfiltrating files and transmitting system information back to its operators, enhancing the group’s surveillance and espionage capabilities.

Cold River—also known in cybersecurity circles as Callisto Group or Seaborgium—has been previously linked to Russia’s Federal Security Service (FSB). The group is notorious for its phishing campaigns and credential theft, often aimed at extracting sensitive information from targets aligned with Western institutions and governments.

Recent Campaigns: Intelligence Gathering on Ukraine and the West

The blog notes that Cold River’s activities have persisted into 2025, with observed campaigns in January, March, and April targeting:

• Current and former advisers to Western governments and militaries
• Journalists
• Think tanks
• Non-governmental organizations (NGOs)
• Unnamed individuals associated with Ukraine

These efforts appear to support Russian strategic objectives, consistent with Cold River’s historical operations.

A Track Record of High-Profile Intrusions

Cold River has been implicated in several high-stakes cyber campaigns over the past few years. In the summer of 2022, the group targeted three U.S. nuclear research laboratories, attempting to gain access to highly sensitive scientific data.

Later that year, in an operation exposed in May 2022, the group published the private email correspondence of former MI6 chief Richard Dearlove and other pro-Brexit figures. The campaign appeared to be part of a broader disinformation effort aimed at influencing public discourse in the United Kingdom.

Image: Freepik

Apple is reportedly developing its own AI-driven search engine, a move that could significantly shift the dynamics of its long-standing relationship with Google. The revelation, reported by Bloomberg, suggests Apple may be looking to reduce or even sever its dependency on Google’s search technology—a partnership that has been a cornerstone of both companies’ mobile strategies for over a decade.

A New Chapter in Search

The news comes amid growing interest in next-generation, AI-powered search engines, such as those developed by Perplexity, OpenAI, and Anthropic. These tools use large language models to provide more conversational, context-aware answers—posing a direct challenge to traditional search engines like Google.

Apple appears to be eyeing a similar path. The company is exploring the integration of artificial intelligence into its search experience, possibly laying the groundwork for a native AI search solution across its ecosystem.

Confirmed in Court

Apple’s intent was disclosed during Google’s ongoing antitrust trial, where Eddy Cue, Apple’s Senior Vice President of Services, testified to the shifting search landscape. Cue noted a recent decline in user searches—a trend he attributed to growing adoption of AI-powered tools that offer faster, more relevant results.

Cue also revealed that Apple is “actively looking at” integrating AI search providers—such as OpenAI and Anthropic—directly into its Safari browser, further signaling a break from its reliance on Google.

This shift could have major financial implications. Google currently pays Apple billions annually to remain the default search engine on iOS devices. If Apple ultimately replaces or deprioritizes Google in favor of its own AI solution, it could disrupt a significant revenue stream tied to Google’s advertising business.

Market Reaction

Investors responded swiftly to the potential shake-up. Shares of Alphabet Inc., Google’s parent company, fell more than 7% following the news, erasing over $75 billion in market value. Apple, by contrast, saw a relatively modest decline of around 2%.

The stark contrast reflects growing investor concern that Google could lose its grip on one of the most valuable digital entry points: the default search position on Apple devices.

The Bigger Picture

While Apple has yet to make any official announcements via a press release or public blog post, its ambitions in AI search align with broader trends reshaping the tech industry. As consumer expectations evolve, driven by advances in generative AI and large language models, traditional web search is increasingly being reimagined.

Should Apple move forward with its own AI-powered search product, the move would not only intensify competition in search—it could also redefine how users interact with information on mobile devices.

Image: Apple Newsroom

In a seismic shift for the mobile app economy, a federal judge has ruled that Apple violated a longstanding injunction by continuing to block app developers from directing users to alternative payment systems—potentially costing the tech giant billions and opening the door for new market entrants.

The decision stems from the high-profile legal battle Epic Games, Inc. v. Apple Inc. (Case No. 21-16506), first launched in 2020 when Epic challenged Apple’s control over in-app transactions. U.S. District Judge Yvonne Gonzalez Rogers, who presided over the original case, found Apple in contempt of the 2021 injunction she previously issued. 

The order prohibited Apple from preventing developers from steering users to non-App Store payment methods.

Apple’s Defiance and Legal Fallout

Despite the clear terms of the original injunction, Apple introduced a new 27% fee on purchases made outside the App Store—a move the court interpreted as a tactic to maintain its dominant revenue model.

In her ruling, Judge Gonzalez Rogers stated:

“Apple, despite knowing its obligations thereunder, thwarted the Injunction’s goals, and continued its anticompetitive conduct solely to maintain its revenue stream.”

The court also referred Apple’s Vice President of Finance, Alex Roman, to the U.S. Attorney’s Office for potentially lying under oath about the fee’s timing—an unprecedented escalation that signals the court’s frustration with Apple’s noncompliance.

Apple has filed for an emergency stay with the Ninth Circuit Court of Appeals, hoping to pause the enforcement of the ruling during the appeals process.

Consequences for Apple—and the Entire Tech Landscape

This decision does more than punish Apple; it redefines how app commerce can function on iOS. The ruling confirms that developers must be allowed to include in-app links or buttons directing users to alternative payment platforms.

Here’s what it could mean going forward:

1. Rise of Third-Party Payment Solutions

The ruling creates fertile ground for a wave of fintech innovation. Startups and established players like Stripe, PayPal, and Adyen are likely to roll out mobile-native payment tools that integrate seamlessly into apps. These solutions could offer lower fees, better user experience, and more flexible payment models, threatening Apple’s dominance.

2. Reduced Costs and New Business Models for Developers

Smaller developers and content creators—long squeezed by Apple’s 15–30% commission—now have a chance to take control of their payment flow. Subscription-based apps, games, and streaming services may lower prices or introduce new service bundles via direct web links.

3. Global Ripple Effects

This case complements regulatory moves abroad, including the EU’s Digital Markets Act and South Korea’s App Store reform. Together, these efforts are reshaping global norms around platform governance and competition.

4. Apple’s Next Move: Compliance or Control?

While Apple must comply with the order, industry analysts expect the company to introduce new terms that retain some leverage—such as additional security vetting, administrative fees, or user interface restrictions for apps using third-party payments.

“This ruling may mark the beginning of a parallel app commerce economy—one where Apple no longer owns the toll booth,”

said Antonia Ramirez, a legal scholar focused on digital markets.

“It’s not just a legal victory for developers—it’s a green light for innovation.”

Looking Ahead

Apple’s appeal may yet delay the practical implementation of the ruling, but the legal foundation has been laid. The mobile app ecosystem—long shaped by Apple’s strict controls—now faces a reckoning. Developers, users, and competitors alike are watching closely as the rules of digital commerce are rewritten in real time.

Image: Freepik

IBM has introduced a suite of new technologies aimed at scaling enterprise AI across hybrid environments, as announced today at its annual THINK conference. This move reinforces IBM’s commitment to breaking down barriers to AI deployment with integrated solutions that unify data, orchestrate AI agents, and simplify enterprise operations.

With over one billion applications projected to emerge by 2028, businesses are under increasing pressure to streamline operations across fragmented systems. IBM is responding with hybrid technologies, enhanced agent capabilities, and the consulting expertise of IBM Consulting, aiming to help clients accelerate AI adoption and realize measurable business value.

Arvind Krishna, Chairman and CEO of IBM

“The era of AI experimentation is over,”

said Arvind Krishna, Chairman and CEO of IBM.

“Today’s competitive advantage comes from purpose-built AI integration that drives measurable business outcomes.”

Enterprise AI Agents Powered by watsonx Orchestrate

Central to IBM’s announcement is the expanded functionality of watsonx Orchestrate, designed to help businesses create, deploy, and manage AI agents across a wide array of enterprise tools.

Key features include:

• Agent Builder: Enables users to create custom AI agents in under five minutes using no-code to pro-code tools.
• Pre-built Domain Agents: Ready-to-use agents tailored for HR, sales, procurement, and more.
• Wide Integration: Supports over 80 leading enterprise applications, including solutions from Adobe, AWS, Microsoft, Oracle, Salesforce Agentforce, SAP, ServiceNow, and Workday.
• Agent Orchestration: Coordinates multi-agent workflows and tools across vendors.
• Agent Observability: Provides monitoring, guardrails, and lifecycle governance.

IBM also unveiled the Agent Catalog, offering access to over 150 agents and tools co-developed with partners like Box, Mastercard, Symplistic.ai, 11x, and others. Sample integrations include a Salesforce-native prospecting agent and a conversational HR agent for Slack.

Tackling Integration Challenges with webMethods Hybrid Integration

A common obstacle to enterprise AI adoption is integration complexity. IBM is addressing this with the launch of webMethods Hybrid Integration, a solution for automating workflows across applications, APIs, partners, and clouds.

According to a Forrester Total Economic Impact (TEI) study, organizations using webMethods realized:

• 176% ROI over three years
• 40% reduction in downtime
• Up to 67% time savings on project execution
• Improved ease of use, security, and visibility

This integration technology complements IBM’s existing automation offerings and is enhanced by collaborations with HashiCorp, including integrations with Terraform and Vault, to support secure, scalable hybrid cloud operations.

Unlocking the Value of Unstructured Data

Unstructured data—such as contracts, spreadsheets, and presentations—represents a largely untapped asset in enterprise AI. IBM is advancing watsonx.data to help organizations activate this data with greater accuracy.

Highlights include:

• Open Data Lakehouse: Now with data fabric features like lineage tracking and governance.
• watsonx.data integration: A unified tool for orchestrating data pipelines across formats.
• watsonx.data intelligence: AI-driven tools for deep insights from unstructured data.

Early testing indicates watsonx.data enables up to 40% more accurate AI compared to traditional retrieval-augmented generation (RAG) methods.

In support of this strategy, IBM recently announced plans to acquire DataStax, a leader in unstructured data handling for generative AI. Additionally, watsonx is now integrated with Meta’s Llama Stack, further enhancing its generative AI capabilities.

IBM’s Content-Aware Storage (CAS) is also now available on IBM Fusion, with support for IBM Storage Scale arriving in Q3. CAS enables contextual processing of unstructured data to speed up AI inferencing.

Infrastructure for AI at Scale: Introducing IBM LinuxONE 5

To support high-performance AI workloads, IBM introduced IBM LinuxONE 5, its most secure and powerful Linux platform to date. It can process up to 450 billion AI inference operations per day.

Innovations include:

• Telum II AI Processor & IBM Spyre Accelerator: High-speed processing for generative AI, with Spyre available in late 2025.
• Confidential Containers & Quantum-Safe Encryption: Advanced security for sensitive workloads.
• Cost and Energy Efficiency: Compared to x86 systems, LinuxONE 5 can lower total cost of ownership by up to 44% over five years.

IBM also expanded partnerships with AMD, Intel, CoreWeave, and NVIDIA to deliver new compute and storage solutions for AI-driven applications.

The Road Ahead

IBM’s latest announcements represent a strategic push toward operationalizing AI at enterprise scale through modular, secure, and hybrid-ready technologies. As businesses move from AI experimentation to enterprise-wide adoption, IBM aims to be the infrastructure backbone enabling that transformation.

Image: IBM video

While many developers have mastered the basics, senior developers should be well-versed in advanced TypeScript concepts to write more maintainable, scalable, and efficient code.

Below are seven advanced TypeScript concepts that every senior front-end developer should understand.

What Every Senior Front-End Developer Should Know About Advanced TypeScript Concepts

1. Union Types

Union types enable a variable to hold one of several specified types, providing the flexibility of dynamic data handling while preserving type safety. It’s akin to having a multi-functional tool that fits diverse scenarios.

Example:

More Real-World Example:

2. Intersection Types

Intersection types combine properties from multiple types, ensuring that a variable satisfies all the conditions. It’s a way to merge the identity of different entities, useful for designing more complex types.

Example:

In this scenario, a SuperAdmin type ensures that a user has both User and Admin properties.

3. Type Guards

Type guards act as validation checkpoints within code, allowing for safer type narrowing during runtime checks. These are like checkpoints that only let specific types through, enhancing security and accuracy.

Example:

4. Conditional Types

Conditional types provide powerful type transformations based on certain conditions. They enable dynamic type manipulation, similar to how logic gates control paths in circuits.

Example:

Conditional types allow developers to build types that adapt based on other types, leading to more expressive and flexible code.

5. Mapped Types

Mapped types transform existing types into new structures by iterating over each property, making it easier to implement type-wide modifications.

Example:

6. Template Literal Types

Template literal types use string literals to create new types, providing a way to enforce naming conventions or combine string patterns for type safety.

Example:

These types make your type definitions more expressive, offering clearer guidelines for code patterns.

7. Recursive Types

Recursive types are crucial for representing complex data structures like trees, linked lists, or deeply nested objects. They allow types to refer to themselves, providing a way to model data structures that are inherently recursive.

Example:

Recursive types are like building blocks that enable you to model structures as intricate as a family tree or a file directory.

Conclusion

Advanced TypeScript concepts such as union types, intersection types, type guards, conditional types, mapped types, template literal types, and recursive types are essential tools in a senior developer’s toolkit. Mastering these features ensures that your code is type-safe, concise, and scalable, setting a strong foundation for complex application development.

Material provided by: Kristiyan Velkov

Photo by Austin Distel on Unsplash

Kristiyan Velkov, Contributor at DevStyleR, is with over 10 years of experience in JavaScript and open-source development. He has honed his skills in tools like React.js, Angular, Vue.js, and Next.js. He also dive deep into DevOps, web accessibility, and security—because great code is about more than just functionality!

LinkedIn         X

Want to know what to expect this year, which project areas are in development, the challenges ahead, and why participating in festivals and events matters so much? You’ll find all that and more in the special video they recorded for DevStyleR.

A fierce and dynamic competition is on the horizon, so make sure to mark April 25 and 26, 2025, in your calendar. That’s when the milestone 10th edition of the FISSION 2025 scientific festival will take place. Young talents will compete in various categories, including biomedicine, ecology, computer science, mathematics, physics, and engineering. There’s even a special category reserved for younger students in grades five through seven.

The video is provided by the FISSION 2025 organizers from the American College of Sofia. 

The virtual DOM is a programming concept where an ideal, or “virtual”, representation of a UI is kept in memory and synced with the “real” DOM by a library such as ReactDOM. This process is called reconciliation.

This process significantly enhances the performance of web apps.

Understanding the virtual DOM is essential for developers who want to get the best out of React. It plays a key role in how React updates the UI, ensuring that changes are applied quickly without unnecessary re-renders.

What Is the Virtual DOM and How Does It Work?

The virtual DOM is an in-memory representation of the real DOM elements.

Instead of interacting directly with the real DOM, which can be slow and costly in terms of performance, React creates a virtual representation of the UI components.

This virtual representation is a lightweight JavaScript object that mirrors the structure of the real DOM.

Here’s a step-by-step process of how the virtual DOM works:

Understanding React’s Rendering and Update Lifecycle

Step 1: Initial Rendering (Virtual DOM Creation)
When the app starts, React builds the entire UI as a Virtual DOM, a lightweight JavaScript representation of the real DOM.

Step 2: Reacting to State and Props Changes
As the app runs, any changes in state or props trigger a re-render of the affected components. React updates the Virtual DOM to reflect these changes, without immediately modifying the real DOM.

Step 3: Diffing Algorithm — Spotting the Differences
React employs a highly efficient diffing algorithm to compare the updated Virtual DOM with its previous version. This comparison identifies only the changes, or “diffs,” between the two versions.

Step 4: Reconciliation — Deciding What to Update
Based on the identified differences, React determines the most efficient way to update the real DOM. Rather than re-rendering the entire UI, React targets only the specific parts that need updating.

Step 5: Applying Updates to the Real DOM
Finally, React updates the real DOM to match the Virtual DOM.

For example, let’s say we have the following counter functionality in the App component:

The virtual DOM representation will look like this:

When the Increase button is clicked once, only the h1 element is changed:

Comparing the Virtual DOM to the Real DOM

The real DOM is a built-in standard interface in browsers that represents and interacts with HTML elements, from Doctype declaration and the root html element to every other element in it.

This real DOM represents the whole HTML document as a tree structure and allows JavaScript to manipulate and change HTML documents.

Sometimes when those changes occur, the whole document might re-render.

This is in contrast to the virtual DOM, which uses a diff algorithm to compare the current and previous versions of updates to the DOM. It only re-renders the parts of the UI that have changed, instead of the whole thing.

Conclusion

The Virtual DOM is one of React’s core features, designed to enhance performance and ensure efficient UI updates. By leveraging the Virtual DOM, React can batch updates, reduce the frequency of expensive reflows and repaints, and apply changes with precision. This streamlined process results in faster, smoother UI updates, significantly improving the overall user experience.

Images and material provided by: Kristiyan Velkov

Kristiyan Velkov, Contributor at DevStyleR, is with over 10 years of experience in JavaScript and open-source development. He has honed his skills in tools like React.js, Angular, Vue.js, and Next.js. He also dive deep into DevOps, web accessibility, and security—because great code is about more than just functionality!

LinkedIn         X

One of the highlights for Sirma in 2024 was the consolidation of six subsidiaries under a unified  brand. This strategic move aimed to optimize operations, expand service offerings, and  strengthen client relationships. The consolidation already has positive financial effects, with  improved processes and increased efficiency leading to expanding business opportunities.

Momchil Zarev, Chief Growth Officer at Sirma, commented on the restructuring process:

“We are working towards providing maximum transparency so that more people can  understand the potential of the business we are developing. Whenever there is a  merger, there are initial structural changes, but that is part of the process.”

Momchil Zarev, Chief Growth Officer at Sirma

As part of its expansion strategy, Sirma acquired a majority stake in the Romanian technology  company Roweb Development. This acquisition not only expands Sirma’s market presence but  also aligns with the company’s ambition to become the leading IT provider in the Balkan region.  The Romanian market, being three times larger than Bulgaria’s, presents significant growth  opportunities. Sirma is actively seeking companies with more than 100 employees in the region,  particularly those, expanding into Western European markets.

Employee development remained a top priority for Sirma in 2024. Over 90 professionals  participated in skill development programs, reflecting the company’s commitment to the  personal and professional growth of its team.

“Unlike many other software companies, we have chosen to invest in our team and  enhance their qualifications at this moment. Sirma sets long-term goals that require  continuous development of high-quality specialists and know-how,”

Zarev added.

Although consolidation is often associated with cost optimization, Sirma has taken a different  approach:

“Almost always, when companies restructure, there are optimizations (in terms of staff).  In our case, these optimizations have been minor – below 1% so far. The goal of the  merger was not about optimization but rather the opposite – to achieve better results with  our existing resources. Our approach was somewhat different – we aimed to generate  business growth rather than just optimization.”

Among the company’s long-term objectives is listing on the Frankfurt Stock Exchange and  expanding into the German market, where Sirma already has significant clients. Positive  feedback from these clients further strengthens the company’s confidence in this direction. The  NASDAQ remains the ultimate long-term goal for Sirma, making its shares accessible to a  broader range of investors.

For Sirma, 2024 was a year of transformation, marked by successful projects, strategic  acquisitions, and a strong commitment to employee development. Moving into 2025, the  company will continue to expand its services in key verticals such as software development,  business and technology consulting for banking, fintech, insurance, transportation, digital  healthcare, logistics, hospitality, e-commerce, public sector, cross-industry, and packaging industries. Sirma’s dedication to innovation remains a driving force in redefining industry  standards and supporting business growth across various fields.

Content and images courtesy of Sirma

Stanislav Tashev is a final-year student at the Professional High School of Computer Programming and Innovation in Burgas, a Microsoft Certified Trainer, an active member of the “Association of Graduates of the Professional High School of Computer Programming and Innovation,” and plays a key role in organizing the upcoming edition of Code Of The Future. His focus is on actively supporting the development of young programmers by sharing his experience and expertise.

We talked with Stoyan and Stanislav Tashev about the upcoming edition of the Code Of The Future 25* hackathon.

How did the idea for the “Code of the Future” hackathon come about?

Stoyan: The idea for the “Code of the Future” hackathon was born from our personal experience as participants in many similar competitions. In the past, we have taken part in hackathons, even winning top prizes, but we were always interested in how such an event is organized from the organizers’ perspective. The best way to find out was to organize a hackathon ourselves. We decided to create “Code of the Future” and offer a platform that combines all the good practices we’ve seen with our own innovative ideas.

How do you determine the themes of the competition?

Stanislav: Our main goal is for the competition’s theme to be a surprise and announced on the day of the event, to ensure fairness and equal conditions for all participants. In last year’s edition, we gave each sponsor the opportunity to propose their own theme if they wished. This year, we’ll make the format even more interesting—sponsors will again be able to suggest themes, but at the opening ceremony, we’ll choose only one at random. This adds an element of surprise not only for the participants but also for us as organizers.

How prepared do participants need to be? What advice would you give them?

Stoyan: The hackathon is open to everyone, regardless of their level of technical preparation. We changed the evaluation criteria to place more emphasis on ideas rather than technical execution. Of course, participants will need to create a prototype (MVP) and validate their idea, but we encourage greater creativity and innovation. Our advice is to spend enough time developing the idea rather than jumping straight into coding. From personal experience, we know the first day is best used for generating the idea and distributing tasks within the team. Everyone can contribute according to their skills—some can do market research while others focus on development.

What is different about the 2025 edition?

Stanislav: Besides the new approach to choosing the theme, this year we’ve listened to the feedback from the previous edition to improve the participants’ experience as much as possible. We have prepared numerous surprises, more conveniences—including more food and coffee—as well as improvements to the mentorship program.

What is your favorite part of organizing the competition?

Stoyan: It’s very hard to pick just one aspect, but one of the most exciting moments is when we first hear the participants’ ideas and have the opportunity to discuss them. It’s a special moment when you see creativity and innovation come to life.

What challenges do participants most frequently encounter, and how do they overcome them?

Stanislav: Participants often face various challenges—from technical problems to a lack of clarity about how to structure their idea. That’s why, as in previous years, we’ll provide mentors. This year, however, we are adding a mentor reservation system so that teams can have personalized time with a mentor who focuses on their idea, presentation, and validation. Mentors won’t write code for them, but they will provide valuable guidance throughout the process.

How does the hackathon influence students’ skills and the projects they develop?

Stoyan: The hackathon is an excellent opportunity to develop skills in innovation, idea validation, and creating an MVP—a minimum viable product. This is a crucial process for starting a successful startup. Participants gain experience in presenting their ideas, analyzing the market, and creating practical solutions.

Can the projects be realistically applied in a business environment?

Stanislav: Absolutely. We even have an example of a project that started as a hackathon idea and is currently being successfully implemented in a business environment as a startup project. This shows that the innovative solutions created during the competition can find real-world applications.

The Silicon Valley giant has released the first two models of the Llama 3 family – one with 8B parameters and one with 70B. According to Meta, these models are significantly better than the Llama 2 models, offering a much lower false rejection rate, improved alignment, and greater diversity in model responses. The specific capabilities of the models such as, code generation and instruction following have also been greatly improved.

Llama 3 was pre-trained on more than 15T tokens from publicly available sources, therefore the training dataset of Llama 3 is seven times larger than the training dataset of Llama 2 and contains four times more code.

According to Meta, in developing Llama 3, a new human evaluation set for benchmarking was also developed that contains 1,800 prompts across 12 use cases. These include asking for advice, classification, answering a closed question, writing code, creative writing, extraction, answering an open question, reasoning, rewriting, summarizing, and more.

The 70B parameter model beat Claude Sonnet, Mistral Medium, GPT 3.5, and Llama 2 using this new assessment set.

Llama 3 is available on AWS, Databricks, Google Cloud, Hugging Face, Kaggle, IBM WatsonX, Microsoft Azure, NVIDIA NIM and Snowflake. In addition, some hardware vendors will also offer support for it, including AMD, AWS, Dell, Intel, NVIDIA and Qualcomm.

Over the next few months, the company plans to update Llama 3 with new features, longer context windows, and larger model sizes.

Read more:
1. GitLab Releases GitLab Duo Chat with 40+ New Features
2. From Competitions in Informatics to MIT – Rumen Hristov’s Formula for Success
3. OpenSSF, CISA and DHS Join Forces in New Open Source Project