Image credits: Pexels
Five tech leaders are using AI to stretch lean teams across bigger workloads.
The advantage of a small team can disappear quickly once administrative work starts eating into the hours employees need for customers and bigger decisions. Across several corners of the tech industry, companies are using AI to reclaim some of that time and stretch the capacity of the people they already have.
The more interesting development is happening beyond standalone productivity tools. Some companies are building AI directly into how work moves through the business, with humans remaining involved where judgment, expertise, and accountability come into play.
Healthcare AI Takes on Complexity
Patient access offers a particularly complicated example. Stephen Dean, co-founder and COO of Keona Health, said the company counted “23 million pathways that an agent can take from when the patient calls,” illustrating how many decisions can sit behind an interaction that appears simple to a patient.
Keona Health uses a deterministic expert system for the logic within its CareDesk CRM, with large language models serving as an interface.
“Everything else has always been not a black box, but traceable,” Dean explained. “Traceable, fixable—you know what’s going on, you know why.”
The company has brought similar technology into its own operations. Dean said its EBITDA margins have increased by roughly 40 points over three years as it applied the same AI principles internally.
Insurance Experiments Get Faster
Insurance teams face a different headache: lengthy processes built around established systems that companies may have little appetite to replace. FurtherAI works as an overlay to existing technology and targets workflows like underwriting and claims.
Fabio Faschi, enterprise sales director at FurtherAI, described the company’s approach as “human in the loop,” with the technology designed to save underwriting teams time. He said some deployments have produced ROI “upwards of 600%” through improvements involving time, accuracy, and labor.
AI can also make experimentation considerably cheaper. Faschi said a team can produce something “directionally able to prove maybe a hypothesis” within seconds and then decide whether further refinement deserves the investment. That changes the cost of asking, “Would this idea actually work?”
Disconnected Data Gets Experience
Sometimes, the biggest obstacle to a small team is information trapped in too many places. BOSS.Tech grew from that frustration after co-founder and CEO Felicite Moorman encountered problems with disconnected business systems, including a missed $300,000 invoice.
“We implemented Salesforce and never got a clean data deliverable,” Moorman said, recalling the experience that helped prompt a different approach. Boss.Tech now integrates business software into a unified data layer designed for AI use.
For Moorman, the result has changed her own capacity. “I used to be able to work between six and nine big strategic partnerships, and now I’m up to 12 to 15 simultaneously.”
Moorman also sees agents taking on enough operational work that smaller organizations can approach growth differently. She added that larger companies can have more organizational concerns slowing adoption.
AI Agents Become Coworkers
Gagan Gujral, founder of Lama Consultants, has taken the lean-team concept literally.
“My structure is one founder plus ten AI agents,” he said, estimating that AI handles about 70% to 80% of his operational work.
Getting there required learning from failure. Gujral kept a 90-day record of problems with his agents and trained an orchestrator agent to monitor the others and correct their code. He said that brought daily agent-management time down from 14 hours to two or three.
His experience also suggests that autonomous systems need time to become useful.
“Agents need a three to six-month period for data ingestion and training before they become effective,” Gujral said.
That timeline complicates the idea of AI as an instant productivity switch.
Governance Becomes Part of Scaling
Speed creates its own problems when automated systems have access to sensitive information or consequential business processes. Ryan McMillen, CEO of RyanTech, describes his company’s approach as governed AI, built around predetermined operating rules and a zero-trust model.
He explains, “When you adopt AI, you’re just turning it on in your organization.”
“These are pre-thought-out plans for how AI should operate,” McMillen adds. RyanTech uses that framework across a 23-person organization supporting over 200,000 users.
McMillen offered purchase-order generation as one example. A process that previously consumed 30 to 40 minutes now happens in seconds, he said, leaving sales staff with more time for customers. Yet he sees the human component as central to the larger competitive advantage: “The companies that implement AI are the companies that give humans AI to make them better.”
Small Teams Find More Room
Across these companies, AI takes very different forms, from deterministic healthcare systems to autonomous agents and governed automation. The common thread is capacity. Work that once consumed hours can move faster, and information that sat scattered across systems may become easier to use.
For lean organizations, that can change what size means. Companies finding their footing with AI are discovering how elastic a small team can become, stretching its time and expertise across workloads that once would have filled considerably more desks.
