L
How AI Automation Powers 60% of Cisco’s 1.8M Support Cases | LangChain Interrupt
@ LangChain
05/24/25
Takeaways
C
Carlos
05/24/25
@ LangChain
Cisco has transformed customer experience with AI agents, achieving over 95% accuracy in support cases.
C
Carlos
05/24/25
@ LangChain
AI-driven automation touches 60% of Cisco's 1.8 million support cases, significantly enhancing efficiency.
C
Carlos
05/24/25
@ LangChain
The framework Cisco uses to validate AI use cases is based on three key criteria that ensure immediate value for customers.
C
Carlos
05/24/25
@ LangChain
Large Language Models (LLMs) are not effective for predictions; combining them with machine learning models enhances enterprise accuracy.
C
Carlos
05/24/25
@ LangChain
Cisco's AI initiatives are backed by a strong data science team, which is critical for the success of their AI applications.
C
Carlos
05/24/25
@ LangChain
Cisco employs a 'supervisor approach' to decompose complex customer queries across specialized AI agents, improving response quality.
C
Carlos
05/24/25
@ LangChain
To avoid common pitfalls in enterprise AI implementation, Cisco emphasizes the importance of aligning technology with people and processes.
C
Carlos
05/24/25
@ LangChain
Cisco's customer experience strategy focuses on maximizing the value customers receive from their investments in Cisco products and services.
C
Carlos
05/24/25
@ LangChain
The typical framework for customer engagement at Cisco is 'land, adopt, expand, and renew', which guides their customer success initiatives.
C
Carlos
05/24/25
@ LangChain
AI is being adopted at Cisco to optimize processes and enhance the customer experience, ensuring better returns for the business.
C
Carlos
05/24/25
@ LangChain
Hyperpersonalization, proactive support, and predictive insights are key components of Cisco's strategy to enhance customer experience.
C
Carlos
05/24/25
@ LangChain
Cisco has been leading the integration of AI in customer experience for over two years, showcasing its effectiveness at industry conferences.
C
Carlos
05/24/25
@ LangChain
Cisco's approach to AI involves defining use case criteria before implementation, ensuring that only valuable use cases are pursued.
C
Carlos
05/24/25
@ LangChain
Cisco has identified that out of 412 potential AI use cases, only five significantly contribute to business outcomes, highlighting the need for focused efforts.
C
Carlos
05/24/25
@ LangChain
The integration of AI in customer support has led to the automation of 60% of low-priority cases, significantly reducing the need for human intervention.
C
Carlos
05/24/25
@ LangChain
Cisco's AI initiatives include deploying predictive insights for renewals and enhancing support through virtual technical engineers.
C
Carlos
05/24/25
@ LangChain
The company leverages both traditional AI and generative AI to provide a seamless customer experience across various platforms.
C
Carlos
05/24/25
@ LangChain
Cisco's AI models are designed to operate in diverse environments, including on-premises, cloud, and hybrid setups, to meet customer needs.
C
Carlos
05/24/25
@ LangChain
The collaboration with LangChain has enabled Cisco to stitch together various AI components, enhancing the overall customer experience.
C
Carlos
05/24/25
@ LangChain
Cisco's AI-driven customer experience aims to provide proactive recommendations, helping customers avoid issues before they arise.
C
Carlos
05/24/25
@ LangChain
The importance of context in customer interactions is emphasized, as it allows for more personalized and effective support.
C
Carlos
05/24/25
@ LangChain
Cisco's AI strategy includes a focus on security and compliance, particularly for customers in regulated industries.
C
Carlos
05/24/25
@ LangChain
The company has developed custom AI models to achieve high accuracy in specific use cases, enhancing overall service delivery.
C
Carlos
05/24/25
@ LangChain
The integration of AI in customer experience is not just about technology; it also involves understanding customer needs and interactions. I need to build an AI to help customers by understanding their needs and reducing their operational burden, which directly correlates with financial results.
C
Carlos
05/24/25
@ LangChain
Over half of my business is recurring revenue, and anything that reduces the time spent on unnecessary tasks translates to higher revenues and better business outcomes.
C
Carlos
05/24/25
@ LangChain
We achieved a 20% reduction in time spent on renewals in less than three weeks, demonstrating immediate business impact and high accuracy in risk recommendations.
C
Carlos
05/24/25
@ LangChain
The supervisor approach decomposes complex customer queries across specialized agents, allowing us to achieve higher accuracy levels in our responses.
C
Carlos
05/24/25
@ LangChain
Combining deterministic machine learning models with probabilistic large language models (LLMs) helps us reach the accuracy levels we target.
C
Carlos
05/24/25
@ LangChain
Defining use cases and metrics before jumping on new tools is crucial for measuring success in AI implementations.
C
Carlos
05/24/25
@ LangChain
Having a dedicated experimentation team allows for faster innovation and the ability to fail fast, which is essential in a rapidly evolving AI landscape.
C
Carlos
05/24/25
@ LangChain
Achieving high accuracy in enterprise use cases is challenging, especially when integrating SQL with LLMs, which do not work well together.
C
Carlos
05/24/25
@ LangChain
We are proposing a full open-source architecture to evolve the industry towards generative AI, addressing the need for agent directories and authentication.