VK Tech Adds Multi-Level Memory to AI Agents in VK AI Space Platform
VK Tech has enhanced its VK AI Space platform by adding multi-level memory capabilities to GenAI agents. The update allows agents to preserve context across interactions, reuse previously prepared materials, and take into account past corrections made by employees.
The company expects the new memory system to cut the proportion of repeatedly solved tasks by 40–50 percent. Memory is organized into four distinct layers that operate together to give agents longer-term awareness while maintaining strict access controls.
Four Memory Layers
- Session memory stores the current dialogue flow and active tasks.
- User memory records an employee’s profile and individual preferences.
- Project memory collects materials, decisions, and rules specific to a given project.
- Agent memory captures the agent’s own working experience, including approaches that produced results and subsequent human adjustments.
With these layers in place, an agent evolves from a short-term digital assistant into a consistent team participant. For example, when preparing a tender response, the agent can draw on the corporate knowledge base, incorporate previous employee edits, and retain both the original data and the revised versions for future use.
The same mechanism applies to incident investigations, project reporting, and product launches. Agents can locate similar past cases, recall the decisions taken, and factor in the reasoning that led to those decisions.
Access to information remains tightly controlled. An agent receives only the data that the collaborating employee is permitted to see. All stored information resides inside the company’s own infrastructure and is never transmitted to external services. Employees can review the contents of any memory layer and delete unnecessary entries at any time.
VK Tech states that the approach will help organizations convert the experience of individual specialists into a durable corporate asset that persists beyond any single conversation and transfers across successive tasks.
Related articles
Designing and Implementing Private Cloud Security Across Geographically Distributed Data Centers
The article shares practical experience in designing and deploying protection for a private cloud hosted across two geographically separated data center sites. The project aimed to achieve service resilience, meet information security requirements, and maintain manageable infrastructure. While theoretical designs often combine standard best practices such as high availability, network segmentation, next-generation firewalls, and dynamic routing, real-world implementation revealed significant conflicts between architectural components. Multiple contractor teams participated, each bringing different visions for network architecture and security controls, leading to complex negotiations and compromises. The process evolved into an engineering puzzle where proposed solutions had to be integrated into a functional, scalable system despite technical limitations and differing priorities. The full piece focuses on the journey from an idealized design to practical deployment through iterative problem-solving rather than deep technical dives.
Efort Qizhi Demonstrates Humanoid Robots Mastering Industrial Tasks Through Universal Technology Base at WAIC 2026
At WAIC 2026 in Shanghai, Efort Qizhi presented its humanoid robots performing complex tasks on uneven terrain, including lifting irregular boulders and installing large photovoltaic panels without prior task-specific coding. The company showcased its HALO human skill collection suit that captures multi-modal data including vision, force, touch and language to distill human operations into robot-understandable representations. Central to the approach is HumanGPT, a world model tailored for embodied intelligence that integrates collected human skills with sensor data to create virtual dynamic environments for robot decision-making. Efort Qizhi also highlighted its full toolchain covering data collection, storage, governance, training, inference and deployment, built around HumanGPT, Dayan data platform, Modou IDE and Openmind OS. Chairman Dr. You Wei introduced the skill iceberg concept, stressing that true industrial value lies in complex capabilities such as multi-robot coordination and real-time process parameter generation rather than simple pick-and-place actions. The system has already logged over 5000 hours of operation on customer production lines, proving transfer from human expertise to reliable machine productivity across manufacturing scenarios.
Google Tests Third-Party App Store Support in Play Store Following Epic Games Antitrust Ruling
Google is actively testing multiple new features in the Play Store, including a dedicated Play Labs experimental section, direct installation of alternative app stores, and an expanded two-line search interface. The most notable change allows users to install competing app marketplaces directly from Google Play, a capability tied to the reinstated 2024 court decision in the Epic Games antitrust case after the companies' settlement was withdrawn. Although the new menu option for third-party stores has been discovered through app modifications, attempts to access it currently result in errors. Play Labs, an experimental area for previewing upcoming store capabilities similar to features already present in Google Search, was also found but remains non-functional. The updated search bar is already operational, supporting longer natural-language queries and an Ask Store prompt to help users describe needs in detail rather than using short app names. APK analysis indicates these developments are still in progress, meaning Google may modify, delay, or cancel any of them before wider release. The overall direction signals that the primary Android app store is preparing to facilitate competition by distributing rival marketplaces.
Spring Security Tutorial Details Refresh Token Implementation to Complement JWT Authentication
A detailed technical guide explains how to enhance JWT-based authentication in Spring Security by introducing refresh tokens for improved security and user experience. The article covers the creation of a RefreshToken entity stored in the database, along with repository and service layers that handle token generation, verification, revocation, and rotation using UUID and expiration timestamps. It discusses security advantages such as short-lived access tokens combined with long-lived refresh tokens hashed via SHA-256 or HMAC-SHA256, while also addressing risks including database theft and XSS attacks mitigated by HttpOnly cookies. Code examples demonstrate updates to AuthService, AuthController, and SecurityConfig to support registration, login, token refresh via cookies, and logout functionality. Additional best practices include automatic token rotation on refresh and session management across multiple devices. The tutorial emphasizes minimizing credential transmission over networks while maintaining stateful token validation.