Prompt Pack
Prompt Pack

Skip-Level

United Limits AI Booking Agents as Databricks Unveils Always On Genie

By ยท 12 Nov 2026

An IT manager, seen from behind, stands at a pale lilac counter and looks at a landscape monitor with three blank canary, ivory, and copper tiles in a shallow arc, against a bright scarlet wall, with a blank keyboard on the counter.
  • On Nov 11, United said it is restricting unauthorized third party AI agents from booking or managing travel on its platforms and through agencies, citing privacy, unintended purchases, and aviation rules.
  • The same day, Databricks announced always on Genie, an autonomous agent that can run in the background and is grounded in Genie Ontology. Databricks says the always on mode is coming soon to preview and beta programs.
  • On Nov 10, Shopify said 30 engineers rebuilt the Shopify merchant app, more than 300 screens plus widgets, an Apple Watch app, and Siri Shortcuts, from React Native to Swift and Kotlin in 12 weeks, with agents writing most of the code.
  • A Nov 10 JetBrains survey writeup, based on 28,945 developers, finds that more than half of employed developers use AI agents in over 60% of their work tasks, and that Python has overtaken JavaScript as the most widely used language, at 65%.

Two things engineering and IT managers should know this Thursday, from who is allowed to let an agent complete a transaction to how a large production app was rebuilt in 12 weeks with agents writing most of the code.

1. United Limits AI Booking Agents as Databricks Unveils Always On Genie

On Nov 11, United published a newsroom statement, "United Restricts Access for Third Party AI Agents," and updated a Digital Trust Center page on the same policy. United says it is restricting access, directly on its digital platforms and indirectly through travel agencies and aggregators, to prohibit use by unauthorized third party personal AI agents or assistants. It points to the risk that sharing account credentials can expose sensitive information, and that agent actions may result in unintended purchases, account changes, or use of MileagePlus miles. It also says automated AI actions have the potential to violate FAA and DOT rules, and gives flight check in as an example because that step requires a human acknowledgment that bags do not contain banned substances. United's help page says people can still use AI to research and plan trips if the tool does not access united.com or the United app, and that even if a customer has given an assistant permission, United will not necessarily let it complete a transaction.

The same day, Databricks announced the second generation of Genie as an always on, autonomous agent for enterprise work. In a Nov 11 press release, the company says employees can give Genie a goal, and that Genie can take ownership of the work, delegate to subagents, track progress, and reach out through a new priority inbox and email when a person needs to make a decision. CEO Ali Ghodsi is quoted saying Genie runs in the background and comes back when it needs a decision, because it knows how the company works, including official numbers and who can see what. Every action, Databricks says, is grounded in Genie Ontology, a layer that captures a company's definitions, metrics, people, and processes. Genie and Genie Ontology are generally available to Databricks customers. The new always on Genie, along with Genie Search and Ontology Manager, is coming soon to preview and beta programs. Databricks says Genie runs inside a customer's existing environment and follows Unity Catalog permissions and Unity Gateway approval rules and budgets, and that every action is auditable.

Why it matters: These two announcements land on the same day and point at the same management question: who is allowed to let an agent act, especially when money, credentials, or customer data are involved. Our read: write down which agents may act in the background, which actions need a human to finish the transaction, and how those permissions will be logged, before a tool starts completing work your policy has not approved. Sources: United newsroom and Databricks

2. Shopify Rebuilt a 300 Screen Merchant App in 12 Weeks, as JetBrains Finds Agents Already Do Most of the Task Load

On Nov 10, Shopify engineering published a post saying 30 engineers rebuilt the Shopify app from React Native to Swift and Kotlin in 12 weeks. The app has more than 300 screens, home and lock screen widgets, an Apple Watch app, complications, and Siri Shortcuts. Shopify says agents wrote most of the code, and that humans focused on architecture, product behavior, and the final judgment about what was good to ship. Engineers worked in pods of two that owned a product area across both platforms, and Shopify says it required human review on every pull request.

The same day, JetBrains published key findings from its 2026 State of Developer Ecosystem report. The survey ran from May to July 2026 and, after cleaning, is based on 28,945 developers across 198 countries and regions. JetBrains says more than half of employed developers use AI agents in over 60% of their work tasks. For the first time, it says, Python has overtaken JavaScript as the most widely used language, at 65%. TypeScript tops JetBrains' Language Promise Index for 2026, followed by Rust and Python.

Why it matters: Shopify is describing a 12 week rewrite of a large production app, not a weekend prototype, and it is equally clear that the speed came from architecture, tight feedback loops, and human review, not from letting agents ship unreviewed code. Our read: if your teams already use agents for a large share of tasks, the constraint to watch is review and verification capacity, not raw code output. Ask how a rewrite of this size would be staffed, checked, and owned on your team today. Sources: Shopify Engineering and JetBrains

Compiled daily with Grok from public sources and edited by the AIDevPulse team.

Get Skip-Level and the daily brief in your inbox: sign up for the AIDevPulse newsletter. For today's developer news, read today's main brief.

Sources:

Recommended AI prompt

Copy this paragraph into ChatGPT, Claude, Gemini, Grok, or whatever AI you use.

Prompt

I manage an engineering or IT team. Using only the two items in this readout, help me draft a one page plan for the next month that does two things: sets a simple permission policy for which AI agents may act in the background and which actions, including purchases or changes to customer records, must be finished by a human, using United's Nov 11 restriction and Databricks' always on Genie announcement as the frame, and outlines how we would staff, review, and verify a large agent assisted rewrite, using Shopify's 12 week Shopify app rebuild and the JetBrains finding that more than half of employed developers already use agents in over 60% of their work tasks. Keep my team's context in mind: [describe your team's size, the agents you already allow, and whether any of them can complete a transaction]. Do not add facts that are not in the readout, and flag anything I should check before sharing.