Multi-agent chat assistant that lets people at Wix ask business data questions in natural language and steers them to the right metrics and dimensions.
Documented by AgentCV from public sources — not submitted by this team.
Not claimed by or endorsed by the team or its organization. Metrics are cited only as the sources state them; absent metrics render as [unknown].
Sources: Solving Data Discovery at Scale: How Wix Uses RAG and Multi-Agent Systems to Find the Right Data Fast (wix.engineering) · The 95% Breakthrough: Ditching RAG for MAG in Wix's AI-Powered Data Discovery (wix.engineering)
Is this your team? Claim it · Ask us to remove it
Recent activity
Version cuts and proof, newest first — the living track record.
Spec sheet
The benchmark fields — designed for comparison across teams.
- Topology
- Other
- Agent count
- [unknown]
- Listed members
- 5
- Platform
- Custom
- Runs on
- Custom ×5
- Industries
- enterprise-ai
- Task kinds
- question-answeringintent-routinganalysis
- Trust tier
- Self-Reported
- Proof entries
- 2
Topology & roster
System wiring
Node details
Typical Other layout — schematic, not verified wiring
HumanHuman operatorHuman gate
- Tool
- Human operator
- Autonomy
- Human-gated
- directs → Agent
BuilderAgent
- Tool
- Agent
- Autonomy
- Runs autonomously
- commits to → Shared workspace
- directs ← Human operator
ResourceShared workspace
- Tool
- Shared workspace
- Autonomy
- Runs autonomously
- commits to ← Agent
Automation story
self-reportedAnna is the chat front door to Wix's data. Instead of searching the semantic layer's many metrics and dimensions, users ask in natural language and are guided to relevant datasets and metrics, then receive a Data Playground link where results can be viewed and edited. Wix reports very positive user feedback and calls it a big step toward data access for everyone at the company.
How a typical Other team handles a task
Typical Other layout — schematic, not verified wiring
Task arrives
Human operator directs Agent.
Agent builds the work
Agent builds the work.
The artifact lands
The artifact lands in Shared workspace: Agent contributes via "commits to".
Human holds the last word
Human operator holds final approval.
Replicate a typical Other setup
Typical Other layout — schematic, not verified wiring
Ingredients
- HumanHuman operator
- BuilderAgent
- ResourceShared workspace
Setup order
- 1.Provision the substrate: Shared workspace.
- 2.Wire Agent: it receives "directs" from Human operator.
- 3.Declare the human gate: Human operator holds final approval.
5 listed members: the members the cited sources name. The total team size is not stated.
Performance metrics
Windowed metrics with provenance. [unknown] means it was not tracked — an honest hole beats an invented figure.
No metrics on record yet.
Token economics
Cost transparency is part of the honesty architecture. [unknown] means it was not tracked — not that it is zero.
Blueprint
Operational DNA — why it works, how it was built, and how it is overseen. Not files for sale; knowledge of the design.
Wix designed Anna to think and behave like an analyst, with each agent playing one specialized part in understanding and refining a request. In the first version (May 2025), retrieval used semantic search on a Vespa vector store that an hourly Airflow job kept indexed, and Cube served as the semantic engine turning structured payloads into SQL. A July 2025 follow-up says Wix moved the asset-selection step from RAG to an approach it calls MAG while leaving the rest of the agent flow unchanged, and reports dimension-selection accuracy rising from 83% to about 95%.
Users see the generated query in the Data Playground and can edit it there. Wix lists open problems: because the assistant keeps no memory between interactions, repeating a question may not give the same result, and preventing premature or partial answers remains hard.
Proof (2)
The team's shared track record — tasks, incidents, lessons, milestones. Per-entry provenance tags are always visible.
- ArtifactJul 9, 2025self-reported
The 95% Breakthrough: Ditching RAG for MAG in Wix's AI-Powered Data Discovery
Public source cited for this curated profile. Published 2025-07-09; accessed by AgentCV 2026-10-20.
https://www.wix.engineering/post/the-95-breakthrough-ditching-rag-for-mag-in-wix-s-ai-powered-data-discovery - ArtifactMay 5, 2025self-reported
Solving Data Discovery at Scale: How Wix Uses RAG and Multi-Agent Systems to Find the Right Data Fast
Public source cited for this curated profile. Published 2025-05-05; accessed by AgentCV 2026-10-20. Lesson stated in this source: Because the assistant keeps no memory between interactions, the same question can return different results over time, which undermines user trust in the answers.
https://www.wix.engineering/post/solving-data-discovery-at-scale-how-wix-uses-rag-and-multi-agent-systems-to-find-the-right-data-fas
Sign in to add a proof entry.
Sign inAttestations (0)
Named third-party statements from people with first-hand experience. Attestations are what separates Peer-Attested from Evidence-Linked.
No attestations yet. Worked with this configuration or agent? Attest to it using the form below — attestations are named third-party statements and are what separates Peer-Attested from Evidence-Linked.
Sign in to attest to this team.
Sign in