Ledger finds hardware implant in wallets sold by reseller, $86M+ stolen

·1 views

Hardware wallet maker Ledger has confirmed that devices sold by authorized reseller CryptoBilis contained an unauthorized hardware implant. The implant captures the 24-word recovery phrase displayed during initial setup and transmits it via an LTE connection. Blockchain investigator Specter estimates losses exceeding $86 million, primarily among buyers in Southeast Asia. Ledger has paused sales through several resellers and marketplaces in the region and advises affected users to transfer assets to a new device.

Read the full story at DEV Community

This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)

OpenClawCash advises developers to test agent payment systems on testnets

OpenClawCash recommends developers thoroughly test payment agents on test networks before deploying them live. Payment logic errors, such as double payments or incorrect address handling, often only appear when real money is involved. The company's agent API offers test environments like Sepolia and Solana devnets that mimic the production network's behavior. Testing on these networks verifies wallet functionality, transaction preflight checks, and server response times without risking real funds.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Developer creates open-source Google Maps lead scraper requiring only Node.js

A developer has released an open-source tool for extracting business lead data from Google Maps. The tool addresses common obstacles like API costs, rate limits, and complex dependencies by using Node.js and Playwright. It scrapes business information including contact details and social media profiles when configured. The solution operates without requiring Docker, Python environments, or paid API keys.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Multi-Tenant AI Serving Suffers Severe Latency from Cache Thrashing

A severe performance bottleneck called KV-cache thrashing is impacting shared vLLM clusters running multiple AI agents. The issue occurs when different, lengthy agent prompts are randomly distributed across GPU workers, causing them to repeatedly evict each other's cached data. This collapses cache hit rates from over 95% to near zero, dramatically increasing initial response latencies from milliseconds to several seconds. The problem is particularly acute for enterprise AI applications using large-context prompts, where slow initial responses break service level agreements. Standard web load balancing techniques worsen the issue, though coordinated routing solutions can mitigate it.

0 comments Read more at DEV Community

ProgrammingDEV Community ·

Forecast: Enterprise AI model deployment requires careful VRAM sizing for 2026

Enterprise teams planning private GPU clusters for 2026 MoE AI models often underestimate required memory. A common error is sizing for only active parameters per token instead of the full resident model and concurrent request caches. This can cause system failures or degraded performance when deploying models like DeepSeek-V4.1-Flash or Kimi K2.5. The article provides a detailed formula and reference table for calculating total VRAM needs. It also compares the costs of in-house clusters versus using commercial AI APIs.

0 comments Read more at DEV Community