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
Passkeys in Production: How One Developer Replaced Passwords with Face ID in Three Days Using FastAPI and Next.js
Yaroslav Morozov details a complete production implementation of Passkeys based on the WebAuthn standard and FIDO2 protocols. The article explains why Passkeys eliminate phishing and database breach risks compared to traditional passwords while providing biometric login via Face ID or Touch ID. It covers database schema design with PostgreSQL tables for credentials and challenges, SQLAlchemy models, and minimal dependencies including py-webauthn on the backend and SimpleWebAuthn on the frontend. Full working code for registration and authentication flows is provided, showing how Passkeys integrate with existing JWT token systems without major refactoring. The guide includes environment configuration, error handling, challenge management with five-minute TTL, and security considerations such as sign count verification to detect cloned keys.
Yandex Disk Files Become Partially Inaccessible After Sasovo Data Center Incident
A detailed user report reveals that approximately one percent of files stored on Yandex Disk are currently unavailable for download following reported incidents at the Sasovo data center. The problems manifest in three distinct states: missing thumbnails with downloadable originals, visible thumbnails with inaccessible originals returning 504 Gateway Time-out errors, and cases where both thumbnails and originals fail to load. Technical analysis using curl requests traced the failures to specific storage nodes such as s418klg.storage.yandex.net, indicating that some data shards may reside in affected infrastructure while others remain operational. Yandex support requested original files for diagnosis but closed the ticket without providing an official explanation or confirming data integrity. The author emphasizes that the issue affects files across both the Photos and Files sections and recommends maintaining offline backups due to the lack of guaranteed availability during data center failures.
Keurig K-Supreme Smart Coffee Maker Generates Nearly 1 TB of Outbound Traffic in Ten Days
A Keurig K-Supreme Smart coffee maker unexpectedly produced around 1008 GB of outgoing traffic over ten days, overwhelming a home UniFi access point while generating only 9.94 GB of inbound data. The device had been placed on a separate network segment, yet the traffic remained largely internal to the home LAN rather than traversing the internet connection. The anomaly was discovered by user Nomad while assisting family members with network maintenance through the UniFi dashboard. After the coffee maker was powered off, the issue could not be reproduced in subsequent testing, and no packet captures were available to determine the content or root cause of the traffic. The model requires internet connectivity for remote control, scheduling, capsule recognition, and automatic reordering of coffee supplies. No similar incidents have been reported by other users, and the manufacturer has not issued any statement regarding the event.
Hash Functions Part 1: Core Properties, Security Requirements and Practical Applications
The article provides a detailed introduction to hash functions, explaining how they map arbitrary-length input to fixed-length output while satisfying three fundamental security properties. It covers preimage resistance, second preimage resistance, and collision resistance, along with the avalanche effect that makes even minor input changes produce unrecognizable output. The text explains why a 256-bit digest is required to achieve 128-bit collision resistance, referencing the birthday paradox and its implications for MD5 and SHA-1. Practical guidance includes using OpenSSL for hashing, applying hashes in commitment schemes, enforcing subresource integrity on web pages, and securely storing passwords with Argon2 and bcrypt. The post emphasizes that hash functions alone do not guarantee integrity without proper transmission of the digest and announces a follow-up on SHA-2 and SHA-3 internals.