# Agents@Work: Benjamin Cox (Rakuten on Building AI Agents at Scale) Source: https://www.youtube.com/watch?v=SpxRLVNDZ-8 Recap page: https://rapidrecap.app/video/SpxRLVNDZ-8 Generated: 2026-02-23T04:34:25.445+00:00 --- ## Quick Overview Benjamin Cox, VP of AI Strategy & Operations at Rakuten Advertising, explains that successfully building AI agents at scale requires a fundamental mindset shift from traditional data science development towards an action-first, human-in-the-loop approach, focusing on automating repetitive workflows and providing agents as co-pilots to augment human capabilities rather than replacing them entirely. **Key Points:** - Successful scaling of AI agents requires a mindset shift away from purely model-building towards workflow automation, treating agents as co-pilots to augment human workers. - Rakuten Advertising serves large advertisers across retail, finance, technology, and apparel, connecting them with large publisher populations online. - The core challenge in scaling AI agents is determining which tasks to delegate to AI versus which should remain human-driven to ensure continuous improvement and maintain high-quality output. - Cox advocates for an action-first approach, using agents to automate repetitive, low-risk, high-impact tasks that humans dislike, rather than attempting full automation of complex processes immediately. - The rapid evolution of LLMs (like GPT) means that relying on outdated AI toolsets or assuming perfect information for model training leads to failure; continuous learning and iteration are essential. - Rakuten has built internal analyst models and training for business process mapping to guide teams on identifying high-value, automatable tasks. - Successful adoption relies on demonstrating tangible value (like increased productivity or reduced manual work) early on, preventing the feeling of being left behind by AI advancements. ![Screenshot at 00:09: Benjamin Cox introduces the core concept of viewing AI agents as taking over entire workflows, contrasting this with the limited scope of traditional automation efforts.](https://ss.rapidrecap.app/screens/SpxRLVNDZ-8/00-00-09.jpg) **Context:** This video features an interview between Daniel Vassilev, Founder & Co-CEO of Relevance AI, and Benjamin Cox, VP of AI Strategy & Operations at Rakuten Advertising. The discussion centers on the practical challenges and strategic approach Rakuten Advertising takes when building and scaling useful AI agents within a large, complex corporate environment, contrasting the traditional data science approach with a more iterative, agent-centric workflow strategy. ## Detailed Analysis Benjamin Cox outlines Rakuten Advertising's strategy for deploying AI agents, emphasizing that success hinges on a cultural shift rather than just technical implementation. He argues against the traditional data science approach of building models based on perfect information, noting that the rapid advancement of LLMs makes that approach obsolete. Instead, Cox advocates for an action-first approach, using agents as co-pilots to automate repetitive, low-value tasks that humans dislike, thereby increasing overall efficiency and quality of life for employees. He notes that Rakuten has developed internal training, such as for business process mapping, to help teams identify where AI can provide the most meaningful, measurable impact. Cox stresses the importance of starting small, proving value quickly with well-defined use cases, and iterating rapidly (fail fast) rather than attempting overly ambitious automation from the outset. He points out that many legacy enterprise software providers promised comprehensive automation that failed, contrasting this with the current capability of LLMs, which facilitate building specialized agents quickly. He concludes that the key to successful adoption is embedding agents where they can augment human intelligence, allowing teams to focus on higher-value strategic work and continuous learning. ### AI Agent Philosophy - View agents as taking over entire workflows, not just single tasks - Agents act as co-pilots to augment human capability, not replace it entirely - Focus on automating repetitive, low-value tasks that humans dislike (e.g., generating reports, deep research). ### Challenges and Mindset Shift - The speed of LLM evolution makes relying on static models obsolete; continuous learning/iteration is necessary - Overcoming the fear of failure is crucial for experimentation - Moving from a pure data science mindset to a workflow-driven mindset. ### Implementation Strategy - Rakuten uses internal training (like process mapping) to identify high-impact areas - Start with small, high-value use cases that can be validated quickly - Avoid promises of full automation initially. ### Measuring Success - Success is measured by tangible value (e.g., time saved, improved quality) and intangible benefits (e.g., sparking curiosity, improving quality of life for employees). ![Screenshot at 00:00: Benjamin Cox during the interview segment, identified by the lower third graphic.](https://ss.rapidrecap.app/screens/SpxRLVNDZ-8/00-00-00.jpg) ![Screenshot at 00:27: A graphic slide appears briefly, showing an isometric illustration of an office space with the text "Agents@Work" and "HOME OF THE AI WORKFORCE."](https://ss.rapidrecap.app/screens/SpxRLVNDZ-8/00-00-27.jpg) ![Screenshot at 00:44: A shot inside the Rakuten Advertising office showing an employee working at a desk with multiple monitors overlooking the city skyline.](https://ss.rapidrecap.app/screens/SpxRLVNDZ-8/00-00-44.jpg) ![Screenshot at 00:47: Title cards identifying the speakers: Benjamin Cox \(VP of AI Strategy & Operations at Rakuten Advertising\) and Daniel Vassilev \(Founder & Co-CEO at Relevance AI\).](https://ss.rapidrecap.app/screens/SpxRLVNDZ-8/00-00-47.jpg) ![Screenshot at 00:09: Benjamin Cox gesturing while explaining the concept of agents handling entire workflows.](https://ss.rapidrecap.app/screens/SpxRLVNDZ-8/00-00-09.jpg)