How AI-native companies turn workflows into operating capability

AI-native companies are transforming how they operate by integrating AI agents into their workflows. These agents are not just assisting with tasks but are now executing entire processes, making workflows more efficient and scalable. OpenAI’s latest Enterprise Signals report highlights that frontier firms—those in the top 10% of AI usage—generate 8.3 times more output tokens per active user compared to typical firms, up from 2.6 times in January. This widening gap underscores a significant shift: leading companies are connecting AI agents to company context and tools, delegating more substantive work, and making successful workflows easier to repeat.

The Frontier Gap: A Deeper Operating Shift

The difference between frontier firms and typical firms is not just in the volume of AI usage but in how deeply AI is integrated into their operations. Frontier firms are leveraging AI agents to handle more complex tasks, from onboarding new employees to managing customer accounts and driving developer ecosystem growth. This integration allows these firms to delegate more substantive work to AI, freeing up human resources for higher-level tasks.

The challenge for leaders is to ensure that these AI-driven workflows are trustworthy, measurable, and continuously improvable. Leaders must also leave room for experimentation, allowing teams to explore use cases whose value may not be immediately apparent.

Basis: Streamlining Onboarding with AI Agents

Onboarding new employees is a process that has traditionally been time-consuming and inconsistent. Basis, a company that builds AI agents for accounting firms, has revolutionized this process. By integrating Codex into their onboarding workflow, Basis has reduced the time it takes to onboard new employees from two hours to just 30 minutes.

On their first day, new employees are greeted by Codex, which introduces them to key company concepts and completes integration setup in the background. Basis has created a company-specific onboarding skill—a reusable set of instructions and resources—that ensures the process is consistent and repeatable. When recurring questions or exceptions arise, HR can update the skill before the next cohort, making onboarding more efficient and easier to improve.

Clay: Managing Scattered Work with Persistent Context

Clay, a company building a self-learning revenue engine for go-to-market teams, faces the challenge of managing critical deal context that is scattered across various platforms like CRM records, email, Slack, and text messages. To address this, Clay has created a persistent workspace and dedicated subagent for every account.

Each subagent reviews primary sources and updates its deal folder overnight. Every morning, a coordinating agent compiles these updates into a short list of priority actions for the sales team. This workflow saves approximately an hour of inbox triage each night and helps sellers focus on the small actions that can add up over a long enterprise sales cycle.

Clay’s approach shows the importance of a consistent structure, a useful refresh cadence, shared evidence, and human judgment at the point of action in scaling evolving work.

Exa Labs: From Opportunity to Tested Action

Exa Labs, which builds web search infrastructure for AI agents, aims to make its search API available wherever developers could use it. Previously, this required developer relations and account teams to monitor repositories and the wider ecosystem, identify promising integrations, gather context, and coordinate work across systems.

Exa has turned this process into a defined workflow for Codex, with clear priorities, access to necessary sources, and human review before anything ships. Codex now monitors for high-priority integration opportunities, gathers relevant context, creates pull requests, runs tests, and prepares weekly updates. This workflow carries an opportunity from signal to tested artifact while reducing handoffs across research, engineering, and communication.

Six Steps to Experiment and Scale AI Workflows

With the frontier gap widening, enterprise leaders need to give employees room to test consequential workflows, measure success, and turn the strongest experiments into repeatable practice. Here are six steps to achieve this:

1. Choose one consequential value surface. Start with an end-to-end workflow where strategic priorities, systems, handoffs, controls, and measurable stakes meet.
2. Define the outcome and how you will measure it. Name the accountable owner, KPI, baseline, and guardrails.
3. Write the agent’s job description. Define what triggers the work, the outcome, required context, tools, permissions, and how persistently the agent should work toward completion.
4. Build the human system around the agent. Put the people closest to the workflow in the design loop and name who owns the business outcome, domain logic, access and controls, adoption, and daily use.
5. Make experimentation visible and reusable. Capture the process and evidence behind what works and package it as skills, Plugins, or shared workspaces.
6. Carry the operating pattern forward. Preserve the context, permissions, evaluations, review points, owners, measures, and enablement that worked, then apply them to the next value surface.

Frequently asked questions

Read next

• What California’s new AI safety bill means for young builders
• OpenAI Ends Cursor Partnership After SpaceX Acquisition

Want to try all of this hands-on? Start with the free Claude Code from Zero course.