Integrate AI agents directly into core workflows like Slack and institutionalize them as the "first line of response." By tagging the agent on every new bug, crash, or request, it provides an initial analysis or pull request that humans can then review, edit, or build upon.

The problem with AI agents isn't getting them to work; it's managing their success. Once deployed, they operate 24/7, generating a high volume of responses and meetings. Your biggest challenge will shift from outreach capacity to your human team's ability to keep up with the AI's constant activity and output.

In regulated industries, AI's value isn't perfect breach detection but efficiently filtering millions of calls to identify a small, ambiguous subset needing human review. This shifts the goal from flawless accuracy to dramatically improving the efficiency and focus of human compliance officers.

AI is not a 'set and forget' solution. An agent's effectiveness directly correlates with the amount of time humans invest in training, iteration, and providing fresh context. Performance will ebb and flow with human oversight, with the best results coming from consistent, hands-on management.

Instead of relying on subjective feedback from account executives, Vercel uses an AI agent to analyze all communications (Gong transcripts, emails, Slack) for lost deals. The bot often uncovers the real reasons for losing (e.g., failure to contact the economic buyer) versus the stated reason (e.g., price).

The effectiveness of agentic AI in complex domains like IT Ops hinges on "context engineering." This involves strategically selecting the right data (logs, metrics) to feed the LLM, preventing garbage-in-garbage-out, reducing costs, and avoiding hallucinations for precise, reliable answers.

AI product quality is highly dependent on infrastructure reliability, which is less stable than traditional cloud services. Jared Palmer's team at Vercel monitored key metrics like 'error-free sessions' in near real-time. This intense, data-driven approach is crucial for building a reliable agentic product, as inference providers frequently drop requests.

Vercel's CTO Malte Ubl suggests a simple method for finding valuable internal automation tasks: ask people, "What do you hate most about your job?" This uncovers tedious work that requires some human judgment, making it a perfect sweet spot for the capabilities of current-generation AI agents.

To improve the quality and accuracy of an AI agent's output, spawn multiple sub-agents with competing or adversarial roles. For example, a code review agent finds bugs, while several "auditor" agents check for false positives, resulting in a more reliable final analysis.