We all know the problem. It is very hard for NVD to keep up with the reported vulnerabilities every day. Several CVEs are in "Awaiting Analysis" state with out even CPE attribution. Our LLM based agents have been collecting the vendor advisory information (such as work arounds etc)
First lets understand what the purpose of RAG is - its to provide right context to LLMs so LLMs can get you the "right answer", from the "right context" provided by you. If the context is not provided by you, the LLM will be free to
May 2, 2024 Generative AI is being used lately for generating phishing emails , including generating hyper personalized ones. How do we recognize such emails ? Generative AI is good at writing emails because it is good at crafting content with given linguistic characteristics. However, Generative AI is good at understanding not
RAG is the lynchpin for several AI application use cases - 1. RAG helps reduce hallucination by giving context to the prompt 2. RAG acts a short and long term memory for agents 3. RAG and the vector db is the long term database for documentation 4. RAG and the
Among all of the security engineering disciplines, vulnerability engineers have probably the most power as it is one of the rare security disciplines where standards exist to calculate severities and compliance standards come with pre defined SLA. With that great power comes great responsibility and if not used well, makes
A LLM agent is essentially a loop on top of LLMs, trying to solve a user problem or answering a question or investigating an open bounded issue. Collaborative Multi agent systems are essential for achieving complex tasks such as capture the flag. Below are some lessons we have learnt while
TLDR ; - Software has become a commodity and will be more so. Customers would be looking for finishing outcomes, not just automation of their tasks. Customers would be demanding decision making systems, not just decision support systems. The outcomes will be micro, not tied to traditional SaaS software workflows (dashboards,
We track our AWS spend closely but we want to spend on it only when humans need to be involved. If everything is normal, and most importantly if the explanation of the abnormalities is normal, we don't want to take a look. If the explanation is abnormal , then
Bootstrapping is a startup is hard, security space is crowded, LLM agents are finicky, AI is expensive. Yet, its been relatively easy for us at Transilience AI to get paying customers, delightful users which keeping our costs minimal. To put in perspective, we have 1 frontend engineer, Muzaffar Hossain , one