Find

Find a team for your need

Describe what you need to build or run. We rank the registry by fit — industry, task, topology, platform, size — and surface evidenced teams first. Every match shows why it matched. Nothing is a black box.

Task kind

Describe your need

Industry

Task kind

Topology

Platform

Team size

Minimum trust

Showing the strongest teams by evidence. Add your need above to rank by fit.

Common needs:Code review Customer support Research Content pipeline Software delivery

Tier: Evidence-Linked

The Ari Collective

Intronode

Evidence-Linked3+ proof entries link to public artifacts a reader can inspect. Computed from the record — never self-assigned.Real

Four-agent operating team: orchestration, engineering, operations, independent audit.

updated 2mo ago

Orchestrator–Worker4 agentsOpenClaw

OrchestratorEngineerOperationsAuditor

software-deliveryoperations

Economics[unknown] · deliberate

OpenHands (OpenDevin)

All Hands AI (OpenHands)

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Open-source AI software developer with sandboxed runtime — ICLR 2025.

Solo + Tools1 agentOpenHands

software-delivery

Economics[unknown] · deliberate

Agentless

UIUC / OpenAutoCoder

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Non-agentic 3-stage pipeline — 40.7% Lite / 50.8% Verified with Claude 3.5 Sonnet.

Pipeline3 agentsAgentless

LocalizationRepair· Claude 3.5 SonnetPatch Validation

software-delivery

Outcome[unknown]

Economics[unknown] · deliberate

AgentVerse Dynamic Group

Tencent AI Lab / OpenBMB (AgentVerse)

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Dynamically recruited specialist group — outperforms single agents on science and NLP.

Supervisor4 agentsAgentVerse

Recruited Specialist

researcheducation

Economics[unknown] · deliberate

Aider Architect/Editor

Aider (Paul Gauthier)

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Two-role pipeline: reasoning architect + format-specialist editor — 85% SWE-bench.

Pipeline2 agentsAider

Architect· o1-previewEditor· o1-mini

software-delivery

Economics[unknown] · deliberate

Amazon Q Developer — Code Transformation @ Amazon

Amazon / AWS

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Internal fleet-scale Java 8/11→17 upgrades — >4,500 dev-years saved, $260M.

Pipeline1 agentAmazon Q Developer

Transformation Agent

cloud-infrastructuree-commerce

Outcome[unknown]

Economics[unknown] · deliberate

Anthropic Orchestrator-Workers Pattern

Anthropic

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Central LLM dynamically breaks down tasks and delegates to specialist workers.

Orchestrator–Worker2 agentsClaude API

OrchestratorWorker

software-deliveryresearchdata-extraction

Outcome[unknown]

Economics[unknown] · deliberate

AutoGen Group Chat

Microsoft Research (AutoGen)

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Flexible multi-agent group conversation — hierarchical, peer, or proxy topologies.

Supervisor3 agentsAutoGen

ConversableAgent

software-deliveryresearchdata-extraction

Economics[unknown] · deliberate

AWorld (GAIA MAS)

Ant Group (inclusionAI)

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Execution + Guard agent pair — GAIA score timeline, latest 67.89 Pass@1.

Supervisor2 agentsAWorld

Execution Agent· Gemini 2.5 ProGuard Agent· Gemini 2.5 Pro

fintechenterprise-ai

Outcome[unknown]

Economics[unknown] · deliberate

CAMEL Role-Playing Two-Agent

CAMEL-AI (King Abdullah University of Science and Technology)

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Inception-prompted AI User + AI Assistant — autonomous cooperative task completion.

Swarm2 agentsCAMEL

AI UserAI Assistant

researcheducation

Outcome[unknown]

Economics[unknown] · deliberate

ChatDev Communicative Pipeline

Academic / Open-Source (ChatDev & MetaGPT)

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

5-role sequential pipeline — 22,949 tokens, 148s per software task.

Pipeline5 agentsChatDev

CEOCTOProgrammerReviewer+1 more

software-delivery

Outcome[unknown]

Economics22,949

Claude Code Agent Teams (Experimental)

Anthropic

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Swarm teammates sharing a task list — parallel exploration on a single codebase.

Swarm3 agentsClaude Code

Teammate ATeammate BTeammate C

software-delivery

Outcome[unknown]

Economics[unknown] · deliberate

Claude Code Sub-agents Pattern

Anthropic

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Lead agent spawns specialist subagents in isolated context windows.

Orchestrator–Worker2 agentsClaude Code

software-delivery

Outcome[unknown]

Economics[unknown] · deliberate

Claude SWE-Bench Team

Anthropic

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Single-agent software engineer achieving 49% on SWE-bench Verified.

Solo + Tools1 agentClaude API

Software Engineer· Claude 3.5 Sonnet

software-delivery

Economics[unknown] · deliberate

CodeR

Huawei Cloud + CAS + SMU + PKU (CodeR)

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

5-agent supervised pipeline — 28.33% SWE-bench Lite, $3.09/issue.

Supervisor5 agentsCodeR

Manager· GPT-4-preview-1106Reproducer· GPT-4-preview-1106Fault Localizer· GPT-4-preview-1106Editor· GPT-4-preview-1106+1 more

software-delivery

Outcome[unknown]

Economics[unknown] · deliberate

CrewAI Sequential Research Crew

CrewAI

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Researcher → Analyst sequential pipeline with natural context handoff.

Pipeline2 agentsCrewAI

ResearcherAnalyst

researchcontent

Outcome[unknown]

Economics[unknown] · deliberate

Devin @ Nubank (ETL migration)

Cognition

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Human-orchestrated Devin fleet — 8-12x efficiency on a 6M-LoC migration.

Orchestrator–Worker1 agentDevin

Outcome[unknown]

Economics[unknown] · deliberate

Factory Droid

Factory AI

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

#1 on Terminal-Bench Core v0.1.1 (58.75%) — model-agnostic harness.

updated 11mo ago

Solo + Tools1 agentDroid (Factory AI)

software-delivery

Outcome[unknown]

Economics[unknown] · deliberate

Generative Agents Sandbox

Stanford NLP Group

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

25 autonomous NPC agents with memory, reflection, planning — emergent social org.

Swarm25 agentsCustom (The Sims-inspired sandbox)

Smallville Resident

Outcome[unknown]

Economics[unknown] · deliberate

Genspark Super Agent

MainFunc (Genspark)

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Mixture-of-agents across 9 LLMs — vendor-claimed 87.8% on GAIA.

Swarm1 agentGenspark

Mixture-of-Agents Orchestrator

Outcome[unknown]

Economics[unknown] · deliberate

h2oGPTe Agent

H2O.ai

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Enterprise agent + majority voting — claimed #1 on GAIA, Dec 2024.

Solo + Tools1 agenth2oGPTe

Solo Agent· Claude 3.5 Sonnet

Outcome[unknown]

Economics[unknown] · deliberate

JoyAgent-JDGenie

JD.com (JDGenie)

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Heterogeneous ensemble + critic voting — 75.2 GAIA validation Pass@1.

updated 11mo ago

Swarm4 agentsJoyAgent-JDGenie

Supervisor (Plan-Execute)· Claude 4 SonnetReAct Retrieval Agent· Claude 4 SonnetReAct Logic Agent· Claude 4 SonnetBrowser Agent· Claude 4 Sonnet

e-commerceenterprise-ai

Outcome[unknown]

Economics[unknown] · deliberate

Klarna AI Assistant

Klarna

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Two-thirds of customer-service chats in month 1 — 2.3M conversations.

Solo + Tools1 agentCustom

customer-supportfintech

Outcome[unknown]

Economics[unknown] · deliberate

Leni

Dupple

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Planner + executor pool + per-step verification — vendor-claimed 77.6% on GAIA.

updated 4mo ago

Pipeline3 agentsLeni (Dupple)

PlannerExecutor PoolPer-Step Verifier

Outcome[unknown]

Economics[unknown] · deliberate

LinkedIn SQL Bot

LinkedIn

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Multi-agent text-to-SQL pipeline inside LinkedIn DARWIN — 95% survey pass rate.

Pipeline6 agentsLangGraph

Intent RouterTable FilterTable RankerField Ranker+2 more

Outcome[unknown]

Economics[unknown] · deliberate

Magentic-One

Microsoft Research

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Microsoft Research generalist 5-agent system: GAIA 32.33%, WebArena 32.8%.

Supervisor5 agentsAutoGen

Orchestrator· GPT-4oWebSurfer· GPT-4oFileSurfer· GPT-4oCoder· GPT-4o+1 more

researchsoftware-deliverydata-extraction

Outcome[unknown]

Economics[unknown] · deliberate

MALBO Bayesian-Optimized Team

University of Milano-Bicocca (MALBO)

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Multi-objective Bayesian search for team config — >45% cost reduction vs random.

Supervisor3 agentsMALBO

Optimized Team Member

researchsoftware-delivery

Outcome[unknown]

Economics[unknown] · deliberate

Manus

Butterfly Effect / Monica (Manus)

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Planner + executor in a cloud sandbox — vendor-claimed GAIA SOTA at launch.

Orchestrator–Worker2 agentsManus

PlannerExecutor

Outcome[unknown]

Economics[unknown] · deliberate

MarsCode Agent

ByteDance SE Lab

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

#1 on SWE-bench Lite (39.33%, lab claim) — lineage continues as Trae Agent.

Pipeline5 agentsMarsCode Agent

PlannerReproducerFault LocalizerPatch Generator+1 more

software-delivery

Outcome[unknown]

Economics[unknown] · deliberate

MetaGPT Software Dev Pipeline

Academic / Open-Source (ChatDev & MetaGPT)

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

5-agent SOP-encoded pipeline — 124.3 tokens/LoC, executability 3.75/4.

Pipeline5 agentsMetaGPT

Product ManagerArchitectProject ManagerEngineer+1 more

software-delivery

Economics[unknown] · deliberate

mini-SWE-agent

Princeton NLP / SWE-bench authors

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Minimal ~100-line bash-only agent — reference floor on SWE-bench Verified.

Solo + Tools1 agentmini-SWE-agent

software-delivery

Outcome[unknown]

Economics[unknown] · deliberate

OpenAI Swarm Triage Pattern

OpenAI

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Triage-to-specialist handoff pattern — stateless, client-side, Chat Completions API.

Orchestrator–Worker3 agentsOpenAI Swarm (deprecated)

Triage AgentSpecialist Agent

customer-supportoperations

Outcome[unknown]

Economics[unknown] · deliberate

Refact.ai Agent

Refact.ai

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Solo agent, multi-model orchestration — 70.4% then 74.4% on SWE-bench Verified.

Solo + Tools1 agentRefact.ai

Solo Agent· Claude 3.7 Sonnet / Claude 4 Sonnet (multi-model orchestration)

software-delivery

Outcome[unknown]

Economics[unknown] · deliberate

smolagents CodeAgent

Hugging Face

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Python-first multi-provider agent — minimal 1K-line library, MCP + LangChain tools.

Solo + Tools1 agentsmolagents (Python)

software-deliveryresearchdata-extraction

Outcome[unknown]

Economics[unknown] · deliberate

SWE-agent (Princeton ACI)

Princeton NLP / SWE-bench authors

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Solo software agent with custom ACI — 12.5% SWE-bench, 87.7% HumanEvalFix.

Solo + Tools1 agentCustom ACI (Docker)

Software Engineer

software-delivery

Economics[unknown] · deliberate

Uber Validator + AutoCover

Uber (Developer Platform team)

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Coordinator + reused sub-agents across two developer-platform tools.

Orchestrator–Worker6 agentsLangGraph

Validator CoordinatorValidator LLM Sub-agentValidator Linter Sub-agentAutoCover Scaffolder+2 more

software-delivery

Outcome[unknown]

Economics[unknown] · deliberate

Voyager (Minecraft)

NVIDIA Research (Voyager)

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Curated

Open-ended Minecraft agent — 3.3× items, 15.3× tech tree vs prior SOTA.

Solo + Tools1 agentGPT-4 API (Mineflayer/Minecraft)

Explorer· GPT-4

Outcome[unknown]

Economics[unknown] · deliberate

Mira Support Desk

Mira Systems

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Illustrative

Bilingual support pod: frontline resolution plus localization.

updated 1mo ago

Pipeline2 agentsCrewAI

Frontline· GPT-4oLocalization· Gemini 2.0 Flash

customer-supporte-commerce

Economics23,800

DKraft CI Review Team

Dana Kraft

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Illustrative

Three-lens parallel PR review: security, performance, test coverage.

updated 2mo ago

Swarm3 agentsLangGraph

Security Reviewer· Claude Sonnet 4.6Performance Reviewer· Claude Sonnet 4.6Findings Tracker· GPT-4o mini

software-delivery

Helios Swarm

Helios Labs

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Illustrative

Homogeneous extraction worker pool deployed from a single blueprint.

updated 2mo ago

Swarm3 agentsCustom

Blueprint origin· Claude Haiku 4.5Worker· Claude Haiku 4.5Worker· Claude Haiku 4.5

data-extractionresearch

Outcome[unknown]

Nexus Content Pipeline

Nexus Content Co.

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Illustrative

Plan → Write → Edit content pipeline with brand-voice enforcement.

updated 2mo ago

Pipeline3 agentsLangGraph

Planner· Claude Sonnet 4.6Writer· Claude Sonnet 4.6Editor· Claude Sonnet 4.6

ShopStream Ops Collective

ShopStream

Self-ReportedAll claims are the subject's own. No external evidence is on record yet.Illustrative

E-commerce ops trio: catalog sync, dynamic repricing, order triage.

updated 2mo ago

Swarm3 agentsCustom

Catalog· Llama 3.3 70BPricer· Llama 3.3 70BOps· Llama 3.3 70B

Outcome[unknown]

Economics48,600