A promising direction for enterprise-level agents! Traversal is an AI SRE agent company that recently raised $120 million in funding. Their goal is to enable software systems to "keep running themselves." AI is leading to more and faster code writing, but production systems are becoming increasingly unreliable—larger codebases, faster releases, and fewer people truly understanding the system. Traditional monitoring and manual troubleshooting are no longer sufficient. Core Product Capabilities: Production World Model™: Real-time construction of a causal graph of the entire production environment, including services, dependencies, code, changes, behavioral baselines, historical events, etc. Causal Search Engine™: Based on causal reasoning, rather than simple correlation, it quickly locates the root cause across multiple hops (5-15 hops or more). Traversal Workers: Proactive AI SRE agents that autonomously intervene in events within Slack/Teams to investigate, analyze, and assist in remediation. Key scenarios include: intelligent alarm triage, root cause analysis (RCA), self-healing, feeding production knowledge back to development (to prevent recurrence), and production support Q&A. It has been adopted by Nubank, Harvey, PepsiCo, DigitalOcean, and numerous Fortune 100 companies. The official figures state that ARR has increased ninefold this year, usage has increased approximately 100 times, and monthly surveys have grown from about 500 at the beginning of the year to nearly 73,000. Some concepts and background information about SRE: I first heard about SRE from a friend who saw Google adopt this system early on. He quickly followed suit at a large company, which led to his rapid promotion and eventual position as CTO. He felt the product was a good fit for his entrepreneurial aspirations, but he no longer wants to start his own business; he's already doing very well at the large company. SRE Origins: 2003: Benjamin Treynor Sloss (also often referred to as Ben Treynor) of Google founded the first SRE team and coined the term. His task at the time was to oversee production system operations, and he proposed the classic definition: "SRE is what happens when you ask a software engineer to design an operations team." At the time, Google's business was growing extremely rapidly, and traditional operations models (relying on manpower and manual operation) could no longer support the scale. Treynor's approach was: - Solve operational problems using software engineering methods - Automate repetitive tasks as much as possible Ensure reliability through code and systems, rather than simply relying on human monitoring This philosophy matured within Google and was widely disseminated throughout the industry in 2016 through the book *Site Reliability Engineering*, prompting other companies (Netflix, LinkedIn, Dropbox, etc.) to follow suit. https://www.(traversal.com)/blog/traversal-series-b-120-million-ai-sre?utm_source=x&utm_medium=anish-social&utm_campaign=series-b-blog