.png&w=256&q=75)
🤓 Latest Submissions
Auditing Dynamic Routing on AMD MI300X
We reproduced a 2025 adverse-weather detection paper on AMD MI300X and measured its core mechanism: the "dynamic" router outputs near-uniform weights for every image (entropy 0.9998/1.0). It costs +41% latency for no mAP gain.We reproduced a 2025 adverse-weather detection paper on AMD MI300X and measured its core mechanism: the "dynamic" router outputs near-uniform weights for every image (entropy 0.9998/1.0). It costs +41% latency for no mAP gain. We reproduced a 2025 adverse-weather detection paper on AMD MI300X and measured its core mechanism: the "dynamic" router outputs near-uniform weights for every image (entropy 0.9998/1.0). It costs +41% latency for no mAP gain. We reproduced a 2025 adverse-weather detection paper on AMD MI300X and measured its core mechanism: the "dynamic" router outputs near-uniform weights for every image (entropy 0.9998/1.0). It costs +41% latency for no mAP gain.
18 Oct 2026
.png&w=828&q=75)
AgentOS:Adversarial Multi-Agent Knowledge Platform
AgentOS is an enterprise-grade Agentic Knowledge Platform designed to transform how organizations interact with their internal knowledge. Instead of relying on public search engines or general-purpose AI models that may hallucinate or use irrelevant information, AgentOS allows users to build secure, domain-specific knowledge workspaces from their own documents. Users can organize documents into projects and collections, upload PDFs, technical documentation, research papers, company policies, manuals, or reports, and selectively choose which collections should be available for answering each query. This ensures every response is grounded only in trusted sources.
13 Jul 2026
.png&w=828&q=75)
Reddit Intelligence Swarm
Reddit Swarm Intelligence is a full-stack, agentic AI system that transforms Reddit into a live, queryable knowledge base. Using BrightData's web scraping infrastructure, it collects posts and comments on-demand across user-defined categories. A multi-agent LangGraph swarm then processes the raw data: a Keyword Extractor identifies the most relevant search terms, a Database Writer archives everything into a structured PostgreSQL schema organized by hierarchical topics, and a Data Synthesizer generates natural-language answers grounded in the scraped content. The Vue.js dashboard offers a rich analytical surface — users can trigger batch scrapes, read AI-generated category summaries, explore topic metrics via bar charts, and dynamically extract subtopics through an on-demand LLM clustering agent that renders the results as an interactive pie chart. Clicking any pie slice fires an intelligent query directly into the Swarm chat, which synthesizes Reddit sentiment and key perspectives on that subtopic in real time.
19 Jun 2026
.png&w=828&q=75)
Reddit Intelligence Swarm
User provides a query to our system, then we extract a predefined number of relevant Reddit posts and comments using BrightData's API, extracting the information to a PostgreSQL database information, then using three agents running locally, Planner, Researcher, Summarizer and Critic we extract relevant information from the data to provide an answer to the user that satisfies their query. The data is extracted from Brightdata's Reddit APIs for Posts and Comments, using filters that extract the data according to recency. This data is stored in PostgreSQL database and queries to extract relevant information are created by planner and researcher, the data is converted to a logical answer by summarizer and critic is used to evaluate the response.
31 May 2026
