# What is an agentic pod?

An agentic pod is a small, permanent team of people and AI agents that owns one part of a business. It pairs a domain expert who knows the work with a builder who can ship agents, under a named sponsor who is accountable for the result.

Unlike a project team, a pod is permanent: it keeps the workflow it automates, accumulates agents over time, and reports one number for what it costs and one for what it saves. Uber's CTO popularized the term in July 2026. A services firm had been using it publicly since April 2025.

Three numbers frame the category:

- Financial pacing reports went from 2 days to 10 minutes after one Uber pod ([Uber's CTO](https://www.linkedin.com/posts/pneppalli_agentic-ai-adoption-is-on-fire-at-uber-and-activity-7480367291851833344-6Lhm))
- Gartner predicts that by 2028 an average global Fortune 500 enterprise will have over 150,000 agents in use, up from less than 15 in 2025 ([Gartner](https://www.gartner.com/en/newsroom/press-releases/2026-04-28-gartner-identifies-six-steps-to-manage-artificial-intelligence-agent-sprawl))
- Organizational factors account for more than 2x the reported AI impact of individual factors, 67% vs 32% ([Microsoft Work Trend Index 2026](https://www.microsoft.com/en-us/worklab/work-trend-index/agents-human-agency-and-the-opportunity-for-every-organization))

Published 2026-07-30. Updated 2026-07-30. By Saleha Iftikhar, SuperOrgs. Canonical: https://www.superorgs.com/agentic-pods

## Where the term came from

Andres Angelani, CEO of the engineering firm Wizeline, was writing about [Agentic Pods in April 2025](https://www.wizeline.ai/the-dawn-of-agentic-pods-a-new-chapter-in-digital-innovation/), and by [June 2025](https://www.wizeline.ai/scaling-ai-for-real-impact-the-power-of-agentic-pods/) had defined them as cross-functional units combining human expertise with AI agents. Fifteen months later, the CTO of a public company reached the same words independently.

"So we created something called Agentic Pods. The idea is simple. We handpicked ~30 of our most AI-proficient engineers and paired each of them with a domain expert from a business function." (Praveen Neppalli Naga, CTO, Uber, July 7, 2026)

Uber ran sixteen pods across sixteen business functions in two months: finance, legal, operations, marketing, customer support, HR, and procurement. [Business Insider covered it](https://www.businessinsider.com/uber-cto-bets-on-agentic-pods-make-ai-more-efficient-2026-7) within 48 hours. Within eight days, Kieran Flanagan, SVP of Agentic GTM at HubSpot, had adopted the model: "From sprawl to system: agentic pods. The concept is simple: pair an AI-proficient engineer with a domain expert, give them a tight window, and ship a working end-to-end workflow." ([The Great AI Sprawl](https://www.kieranflanagan.io/p/the-great-ai-sprawl), July 17, 2026)

Two parties converging on the same name, fifteen months apart, is better evidence that the name is natural than one party inventing it.

## What is inside a pod

A pod is three roles and a roster of agents. The roles are stable; the agents accumulate.

- **The sponsor.** A leader who answers for the pod by name: owns the outcome, approves what agents may do, and is the escalation path when one gets something wrong.
- **The builder.** The person who can ship agents. Builders are the scarce resource; a pod's build capacity is set by its builders.
- **The domain expert.** The person who does the work today. Per Uber's CTO: "You can't automate them effectively by looking at process diagrams or documentation. You have to understand how the work actually gets done."
- **The agents.** Members, not tools. Each has a name, an owner, and a place on the roster; its cost and output are part of the pod's scoreboard.

## How a pod works: the first ten days

Uber's ten-day shape: days 1-2 shadow the expert and document every step; day 3 prioritize by scale, repetition, business impact, and data availability; days 4-5 build a working agent alongside the person who does the job; days 6-9 validate with others doing the same work; day 10 ship.

Once standing, HubSpot's loop: observe, identify, build, deploy small, expand, enable, measure. Flanagan: "Building it is maybe 40% of the job. Getting adoption is the other 60%."

## The two meanings of "agentic pod"

**Sense one: people plus agents.** The sense on this page. Uber, HubSpot, BCG, and the services firms all mean a unit that contains humans.

**Sense two: agents only.** A cluster of software agents coordinating with each other, no human inside. [Daniel Vaughan's five-role pipeline](https://codex.danielvaughan.com/2026/03/28/agentic-pod-in-practice-multi-agent-roles/) is the clearest example ("The Orchestrator is the only role that talks directly to the user. Every other role is internal to the pipeline."). QualiZeal describes similar agents-only quality pods. A real pattern, honestly used, but it is a multi-agent architecture, and architecture already has better names for it. The person who popularized "agentic pod" put a human domain expert inside the unit by definition, and that pairing is why his pods worked.

## What an agentic pod is not

- **Not a department.** A department groups people by function and owns headcount. A pod crosses functions and owns one workflow end to end.
- **Not a squad or project team.** Project teams disband and the knowledge disperses. A pod is permanent, and every agent it ships makes the next one cheaper.
- **Not a pod you buy.** Globant sells AI Pods by monthly subscription ("Clients can subscribe monthly to an AI Pod, and each subscription includes a token-based metered capacity," [per their launch release](https://www.prnewswire.com/news-releases/globant-launches-ai-pods-a-new-subscription-model-to-reinvent-how-companies-scale-ai-302473988.html)). Wizeline, Infosys, and HatchWorks sell adjacent units. Real services, different thing: when the contract ends, the knowledge leaves with the vendor.
- **Not a Kubernetes pod.** A [Kubernetes pod](https://kubernetes.io/docs/concepts/workloads/pods/) is a group of containers running together on a cluster. The two words are unrelated.

| | Agentic pod | Department | Project team | A pod you buy |
|---|---|---|---|---|
| Who is in it | People and AI agents: sponsor, builder, expert | People, by function | People, per deliverable | A vendor's staff and their agents |
| How long it lasts | Permanent | Permanent | Until the project ships | Until the contract ends |
| What it owns | One workflow, end to end | A function and headcount | A deliverable | A statement of work |
| Who is accountable | The sponsor, by name | The department head | The project lead | The vendor's account lead |
| How it is measured | Spend, savings, capacity, backlog | Budget and headcount | Scope and schedule | Hours or outcomes billed |
| What happens to what it learns | Stays, compounds into more agents | Stays, in people | Disperses at disband | Leaves with the vendor |

## What a pod is measured on

Four numbers, whatever tooling you use: **spend** (what the pod's agents cost this month, metered), **savings** (what the workflow cost before), **capacity** (how much the pod can take on, expressed as a date: "2 of 4 build slots used, next opens Aug 11"), and **backlog** (what is queued). A pod that cannot produce those four numbers is a project with a nice name.

## Where pods fail

- A pod with a backlog and no builder is a queue. The fix is supply: add a builder or hire one.
- A pod with no named sponsor is ownerless. [Gregory Scott Henson](https://www.gregoryscotthenson.com/post/put-your-ai-agents-on-the-org-chart-then-name-the-owner): "An agent with no owner is not an agent. It is a science project."
- A pod run as a project disbands, and the workflow knowledge leaves with it.
- A pod measured on activity produces demos. The scoreboard is spend against savings on an owned workflow.

## Do agentic pods actually work?

Uber's numbers: capital allocation across 150 cities from 15 hours to 30 minutes; financial pacing reports from two days to ten minutes; marketing web quality assurance from two weeks to fifty minutes; 9,000 manually built support workflows replaced with self-service automation.

The objection, from inside the same company: Uber President and COO Andrew Macdonald, [May 2026](https://futurism.com/artificial-intelligence/uber-ai-costs-arent-worth-it), said it is "very hard to draw a line" from AI productivity metrics to "actually producing 25 percent more useful consumer features," and "that trade becomes harder to justify because it's not free. AI is not free."

Both are true, and the gap is a measurement problem. At company level the line is too long to draw. At pod level it is short: a pod owns one workflow, and a workflow has a before and an after. The pod is the smallest unit you can put a cost, a saving, and a date on.

## Who runs agentic pods today

Internally: Uber (sixteen pods), HubSpot (Flanagan's GTM loop), and BCG describes [agentic-marketer pods](https://www.bcg.com/publications/2026/agent-native-marketing-operating-model) of three to five people. McKinsey's technology practice reports [early implementations where teams of eight to twelve give way to smaller pods](https://www.mckinsey.com/capabilities/tech-and-ai/our-insights/rewiring-software-delivery-for-the-agentic-era) supervising agent-driven execution.

Selling them: [Wizeline](https://www.wizeline.ai/agentic-pods/), [Globant](https://www.globant.com/ai-pods), [Infosys](https://www.infosys.com/iki/perspectives/human-agent-collaboration-scale.html), [HatchWorks](https://hatchworks.com/blog/gendd/gendd-pod/).

GitHub and Microsoft ship the same idea as [Squad](https://commandline.microsoft.com/squad-github-copilot-agent-teams-architecture-durable-memory/), "human-led agentic teams": the system "can delegate, parallelize, and preserve context. It shouldn't launder accountability."

## Running pods as software

Everything above works with a spreadsheet and discipline. A platform adds the scoreboard nobody maintains by hand: live spend and savings per pod, capacity computed from the actual builders and experts, agents that keep their owner as the org changes. That is what [SuperOrgs](https://www.superorgs.com) does: pods with faces and a sponsor ring, capacity as a date, and two doors when a pod stalls (add a builder, or hire one through the platform).

## FAQ

**What is the difference between an agentic pod and a squad?** A squad is an agile team shape, usually assembled around a deliverable. An agentic pod is permanent, includes AI agents as members, and owns a business workflow. GitHub and Microsoft ship a similar humans-plus-agents idea under the name Squad.

**How many people are in an agentic pod?** Small. Uber paired one engineer with one domain expert. BCG describes three to five people. Past six or seven you have a team with a backlog, not a pod with an owner.

**Are AI agents members of the pod, or tools the people use?** Members. Each has a name, an owner, and a roster place; its cost and output are on the pod's scoreboard.

**How long does a pod take to ship something?** Uber's pods shipped their first working agent in ten days. Two weeks is a reasonable first cycle.

**Who is accountable when an agent in a pod gets something wrong?** The sponsor, by name. Every agent has a human owner and a reporting line.

**Is an agentic pod the same as a Kubernetes pod?** No. A Kubernetes pod is container infrastructure. The two words are unrelated.

**Can I buy an agentic pod?** You can buy pod-shaped delivery capacity from Globant, Wizeline, Infosys, or HatchWorks. What you cannot buy is the compounding: when the contract ends, the workflow knowledge leaves with the vendor.

**Do agentic pods actually work?** Uber's workflow numbers say yes; Uber's COO says the company-level line to customer value is still hard to draw. Pods work where you measure them: the workflow.

**How do you measure an agentic pod?** Spend, savings, capacity (as a date), and backlog.

**Do I need software to run agentic pods?** No. What gets hard at scale is the scoreboard: live spend and savings per pod, capacity dates, and agents that keep their owners. That bookkeeping is what platforms are for.