Airchat (airchat-cli)

Airchat is Airbnb's internal agentic-coding harness, built by its Dev AI team as a wrapper over vendor coding agents such as Claude Code and Codex. The team first built its own orchestrator from scratch, never shipped it, and delegated to Airchat with a thin shim instead.

Company: Airbnb
Approach type: Platform
Work supported: Coding, Code review
Interfaces: Cli, Web
Deployment stage: Scaled
Evidence strength: Mixed
Entry reviewed: 2026-09-16

Copy as MD Ask ChatGPT

Capabilities and architecture

Model

Vendor coding agents (Claude Code and Codex are named in use), wrapped by Airbnb^{[Source 2]}

Harness

Wrapper over vendor coding agents with a unified gateway, an internal plugin marketplace, and AirDev parallel workspaces^{[Source 2]}

Sandbox

AirDev remote workspaces run agent sessions network-isolated from production services and user data^{[Source 2]}Observation date: 2026

Tool access

More than a dozen internal MCP servers connect agents to internal systems^{[Source 2, Source 3]}

Knowledge

Not reportedThe captures describe internal context through MCP servers and configuration files, not a separate knowledge store.

Context management

AirChat loads default MCP servers into sessions, and the agent loop reads configuration files such as AGENTS.md and CLAUDE.md^{[Source 2, Source 3]}Observation date: 2026

Credentials

AirChat handles authentication and permissioning for internal agent use^{[Source 2]}Observation date: 2026

Interfaces

The CLI is the stated core surface and the Remote web UI is a live internal EAP; AirChat began as JetBrains and Xcode IDE plugins, which the captures describe as history.

^{[Source 2]}

Observation date: 2026

Documented uses

Engineer-written spec to reviewed pull request on the AirChat platform.

Start an agentic coding session

An engineer writes a short spec or prompt that starts an autonomous coding session through AirChat^{[Source 2, Source 3]}

Observed in October 2025

Materialize the code change

The session calls LLMs and internal tools in a loop until it produces a complete code change^{[Source 3]}

Observed in October 2025

Review every line before merge

The loop ends at the diff; a human reviews every line of the change before it merges as a pull request^{[Source 3]}

Observed in October 2025

Access and controls

See documented credential and access boundaries in the architecture above. Scoped human-review assessments for individual uses remain in research details.

Reliability and validation

Reported: Review every line before merge

Adoption and operating evidence

These results come from the sources. Notes below each result explain what it covers and any reported limits.

Headline claim

About 64% of pull requests materialized through agentic coding^{[Source 3]}

Reported by: Airbnb
Scope: Airbnb PRs materialized through agentic coding by the October 2025 talk
Denominator: Airbnb pull requests; exact count not supplied

Observed in October 2025

About this result

Category: Adoption output
Evidence type: Reported measurement
What it covers: Share of Airbnb pull requests materialized through agentic coding by the October 2025 talk

Lessons

Reported opinion: Wrap vendor coding agents with a thin shim instead of building a full orchestrator from scratch; Airbnb's own orchestrator never shipped^{[Source 3]}

Observed in October 2025

Back to infrastructure

Improve this record

Give your coding agent the context and instructions to prepare a source-backed update.

Suggest on GitHub

Paste the prompt into your coding agent, then describe the correction or add a public source.

Preview prompt

Help me update the Internal Agents Map record for Airchat (airchat-cli) at Airbnb.

Repository: https://github.com/steel-experiments/internal-agents-map
Record: data/agents/airbnb-airchat.yaml
Published record: https://internal-agents.com/agents/airbnb-airchat
Readable context: https://internal-agents.com/agents/airbnb-airchat.md

Use the correction or public source I provide in this conversation. If I have not supplied either, ask what I want changed before editing.

Use an existing checkout of this repository, or clone it if needed. Read the current record from the repository. Follow AGENTS.md, CONTRIBUTING.md, and the relevant instructions in data/schema.md.

Verify that the evidence supports the requested change. Preserve uncertainty and distinguish reported facts from catalog judgments. Follow the repository process for capturing new sources. If evidence is insufficient, explain the gap instead of inventing details.

Make a focused change on a branch. Regenerate derived files and run the required checks. Do not edit generated content by hand.

Show me the change, supporting evidence, and check results, including any checks you could not run. Ask whether to submit a pull request unless I already requested one. Do not merge the pull request. If you cannot edit the repository, prepare a source-backed issue draft.

Sources and research details

Citations link to the original publisher. Each source also keeps a preserved copy in the repository, so a changed or removed page stays checkable.

Agentic coding at Airbnb (DPE.org)https://dpe.org/sessions/szczepan-faber-mike-nakhimovich/agentic-coding-at-airbnb/Talk · Direct participant · Last source verification: 2026-08-31
Preserved copy
Beyond the CLI (DX podcast)https://getdx.com/podcast/beyond-the-cli-agentic-ai-for-async-workloads-and-non-developers/Podcast · Direct participant · Last source verification: 2026-08-31
Preserved copy
How to get your team past the AI (The AI Thinker)https://www.theaithinker.com/p/how-to-get-your-team-past-the-aiNews · Independent secondary · Last source verification: 2026-08-31
Preserved copy

Question coverage and scope

purpose: Reported
workflow: Reported
human involvement: Not applicable: There is no single platform-wide agent review boundary. Scoped downstream operating-model claims remain in research details; access controls are described separately.
implementation: Reported
validation: Reported: The documented check is the human line-by-line diff review before merge; no separate automated gate is described for coding runs.
observations: Reported
lessons: Reported

Research details for every claim on this page

Summary
Statement type
Fact
Provenance
Reported
Confidence
Medium
Confidence reason
The build is described by Airbnb engineers in talks and podcasts, not in a first-party engineering blog.
- ContextualizesAgentic coding at Airbnb (DPE.org)
- SupportsBeyond the CLI (DX podcast)
- SupportsHow to get your team past the AI (The AI Thinker)
Headline claim
Statement type
Metric
Provenance
Reported
Confidence
Low
Confidence reason
The 64% figure comes from a third-party newsletter that quotes the engineers, not from a first-party Airbnb source.
Reported by
Airbnb
Scope
Airbnb PRs materialized through agentic coding by the October 2025 talk
Denominator
Airbnb pull requests; exact count not supplied
Method
Not reported
Observation date
2025-10
- SupportsHow to get your team past the AI (The AI Thinker)Preserved content.md, lines 16, 24, 194
Harness
Statement type
Fact
Provenance
Reported
Confidence
Not assessed
Confidence reason
No explicit confidence assessment is recorded for this claim.
- SupportsBeyond the CLI (DX podcast)
Model
Statement type
Fact
Provenance
Reported
Confidence
Not assessed
Confidence reason
No explicit confidence assessment is recorded for this claim.
- SupportsBeyond the CLI (DX podcast)Preserved content.md, lines 114, 214
Tool access
Statement type
Fact
Provenance
Reported
Confidence
Not assessed
Confidence reason
No explicit confidence assessment is recorded for this claim.
- SupportsBeyond the CLI (DX podcast)
- SupportsHow to get your team past the AI (The AI Thinker)Preserved content.md, line 166
Interfaces
Statement type
Fact
Provenance
Reported
Confidence
Medium
Confidence reason
The CLI is the stated core abstraction and the Remote web UI is live as an internal early-access program.
Observation date
2026
- SupportsBeyond the CLI (DX podcast)Preserved content.md, lines 188–192, 214
Sandbox
Statement type
Fact
Provenance
Reported
Confidence
Medium
Confidence reason
An Airbnb engineering manager described the workspace isolation in the podcast transcript.
Observation date
2026
- SupportsBeyond the CLI (DX podcast)Preserved content.md, line 232
Credentials
Statement type
Fact
Provenance
Reported
Confidence
Medium
Confidence reason
An Airbnb product lead stated that AirChat handles authentication and permissioning.
Observation date
2026
- SupportsBeyond the CLI (DX podcast)Preserved content.md, line 118
Context management
Statement type
Fact
Provenance
Reported
Confidence
Medium
Confidence reason
Both direct-participant captures describe default MCP loading and configuration files the loop reads.
Observation date
2026
- SupportsBeyond the CLI (DX podcast)Preserved content.md, line 118
- SupportsHow to get your team past the AI (The AI Thinker)Preserved content.md, line 82
Start an agentic coding session
Statement type
Fact
Provenance
Reported
Confidence
Medium
Confidence reason
The reviewed talk defines the agentic loop as starting from an engineer-written spec or prompt.
Observation date
2025-10
- SupportsBeyond the CLI (DX podcast)Preserved content.md, line 118
- SupportsHow to get your team past the AI (The AI Thinker)Preserved content.md, lines 82, 104
Materialize the code change
Statement type
Fact
Provenance
Reported
Confidence
Medium
Confidence reason
The reviewed talk's loop diagram states repeated LLM and tool calls produce the materialized change.
Observation date
2025-10
- SupportsHow to get your team past the AI (The AI Thinker)Preserved content.md, line 82
Review every line before merge
Statement type
Fact
Provenance
Reported
Confidence
High
Confidence reason
A speaker quote states that every line is reviewed by a human before merge.
Observation date
2025-10
- SupportsHow to get your team past the AI (The AI Thinker)Preserved content.md, lines 178–182
Key observation
About 64% of pull requests materialized through agentic coding
Duplicate representation of Headline claim. Identical text, scope, period, denominator, and confidence as the headline; adds nothing the canonical lacks.
Statement type
Metric
Provenance
Reported
Confidence
Low
Confidence reason
The figure comes from a third-party newsletter, not a first-party Airbnb source.
Reported by
Airbnb
Scope
Airbnb PRs materialized through agentic coding by the October 2025 talk
Denominator
Airbnb pull requests; exact count not supplied
Method
Not reported
Observation date
2025-10
- SupportsHow to get your team past the AI (The AI Thinker)Preserved content.md, lines 16, 24, 194
Lesson
Statement type
Opinion
Provenance
Reported
Confidence
Medium
Confidence reason
A direct participant described the pivot from an unshipped orchestrator to a thin shim as the team's choice, not a measured comparison.
Observation date
2025-10
- SupportsHow to get your team past the AI (The AI Thinker)Preserved content.md, lines 202–210
Operating model assessment
Level 3 for coding task → reviewed pull request; human attention boundary: work-product review.
Statement type
Inference
Provenance
Catalog judgment
Confidence
Medium
Confidence reason
Airbnb engineers describe agents producing pull requests that engineers review, which locates human attention at work-product review.
Observation date
2025
- ContextualizesBeyond the CLI (DX podcast)
- SupportsHow to get your team past the AI (The AI Thinker)Preserved content.md, lines 82, 250

Capabilities and architecture

Documented uses

Access and controls

Reliability and validation

Adoption and operating evidence

Lessons

Related reading

Agents using this

Sources and research details