Skip to content
OpenCorvus
EN中
← Blog

How expert squads share a complex task

A detailed journey through data, training, product, paper and delivery.

Squads, combined

The longest work is not one team working longer. It is several, each owning a stage, each handing the next one something it can read.

A Mission records which Expert Squad IDs are available when it starts; capabilities installed later do not silently widen that set. Each child Task then resolves one admitted ID to one exact package revision plus its selected workflow, fixed for that Task’s lifetime. Composition happens at the Mission level and ownership stays at the Task level.

Case: from research sources to a submitted paper

This is an example division of research responsibilities; the current squad catalog owns role and workflow definitions.

StageExpert SquadHands on
01FrameScientific Research DesignEvidence landscape, competing hypotheses, and a rigor-and-ethics read, joined into one decision register.
02SourceDeep ResearchMulti-perspective discovery and curated evidence, with an independent citation review between the draft and the report.
03AnalyzeData Analysis & Business InsightsMetric reconciliation and parallel performance and segment work, checked by a role that did not run the analysis.
04DraftResearch StudioDurable evidence collection, reproducible analysis, post-computation fact-checking, and template-driven delivery.
05ReviewAcademic Paper ReviewLiterature, novelty, logic, methods and figures — plus a citation-and-hallucination auditor separate from all of them.
06PackageOffice DeliveryThe submission deck built from the same sources, with a real chart and a validation receipt.

Prior-art evidence, live-page observation, or a second language extend the same chain. Adding Patent Landscape and Prior Art, Browser Research & Acceptance, and Localization & Adaptation extends the responsibilities to suit the task.

Read the shape rather than the individual stages. Four of the six squads in that chain carry a role whose entire job is to disbelieve work it did not do — Deep Research’s citation reviewer, Data Analysis’s fact-checker, Research Studio’s own fact-checker, and Academic Paper Review’s citation-and-hallucination auditor. This separates responsibility for delivery from responsibility for review.

Other combinations that already ship

CombinationChain
Deal due diligenceMergers and Acquisitions Due Diligence → Forensic Accounting Investigations → Commercial Legal → Tax Compliance → Internal Audit Control Assurance
Incident to written knowledgeService Reliability Incident Operations → Digital Forensics Incident Investigation → Review & Debug → Knowledge Base Operations
Launching somethingProduct Management → Marketing & Growth Strategy → SEO & Generative Engine Optimization → Product Video Production → Localization & Adaptation

Split where a delivery can be independently owned, accepted, or depended on. Splitting for its own sake produces coordination overhead with no owner. See Squad composition.

Case

Turn DeBERTa into a complete research program

Model, data, CUDA experiments, a live product, figures, paper, review, and repository — one Mission, with a complete squad owning every stage.

6squads44named roles

Original Mission requirements

  1. Acquire DeBERTa v3 Base ABSA v1.1, re-investigate current sources, and search for or synthesize traceable training data before training.
  2. Provision a complete CUDA-only runtime with no CPU training path; record train/test performance for every innovative design and iteration; build an auto-updating training monitor and inference website; keep improving the model against an explicit baseline.
  3. Use the best experiment's exact model architecture and design rationale to produce reproducible, publication-ready figures.
  4. Research related literature and write a complete ACL-style short paper of at least four pages.
  5. Deeply review and proofread the manuscript, eliminate factual errors, and make its organization concise and informative.
  6. Create a well-organized Git repository, execute the work through Mission decomposition, and deliver a reviewed GitHub push.

Required final deliveries

  • Traceable ABSA dataset
  • CUDA-only training runtime
  • Best model checkpoint
  • Live monitor & inference site
  • Architecture & experiment figures
  • Reviewed ACL paper
  • Published Mission repository
View the audited CUDA experiment artifact on GitHub

Six high-level deliveries

  1. 01Model & dataDeep ResearchA verified DeBERTa v3 Base ABSA v1.1 source, current ABSA evidence, and a sourced plan to find, clean, or synthesize training data.
  2. 02CUDA training systemAdvancedA CUDA-only runtime, iterative baseline and candidate training, an experiment ledger, and a live training-monitor and inference website.
  3. 03Architecture evidenceData Analysis & Business InsightsBest-run comparisons, architecture and design diagrams, and reproducible figures bound to the exact winning checkpoint.
  4. 04ACL short paperResearch StudioA concise, informative ACL-style short paper of at least four pages, grounded in related work and the best experiment.
  5. 05Independent paper reviewAcademic Paper ReviewResolved findings across facts, citations, novelty, method, structure, figures, hallucination risk, concision, and informativeness.
  6. 06Mission repositoryBaseA reproducible, organized Git repository with the Mission stage map, reviewed documentation, and a verified GitHub push.
Explore the workflow detailsCollapse the workflow detailsTurn six high-level deliveries into five workstreams, eighteen squad-owned stages, and visible milestones.

Workload ×3

This is the complete Mission OpenCorvus has to coordinate

It starts with model and data evidence, crosses CUDA training, a live product, and research publication, then converges on independent reproduction and GitHub release.

18squad-owned stages99named roles

Model & data evidence

Prove what can be trained, where the data came from, and how improvement will be judged.

3 stages15 named roles

  1. Verify model & literatureDeep ResearchSource-verified model identity, licence, checkpoints, and an evidence map of current ABSA work.
  2. Build the data lineData Engineering ReliabilityVersioned acquisition, cleaning, splitting, synthesis, lineage, and reproducible dataset builds.
  3. Freeze the baselineAI Model Governance and EvaluationA model-and-data card, fixed evaluation slices, quality risks, and a signed-off baseline protocol.
CUDA training & experiments

Build the GPU runtime, implement the trainer, run candidate campaigns, and reconcile every metric.

4 stages32 named roles

  1. Design CUDA runtimeCloud Platform ArchitecturePinned driver, CUDA, framework, container, storage, and GPU observability contracts with no CPU training path.
  2. Implement trainingAdvancedA reproducible trainer, configuration surface, checkpoint lifecycle, and focused verification.
  3. Run candidate campaignsEvolution LabFrozen arms, budgets, mutations, train/test metrics, integrity review, and a best-candidate decision.
  4. Reconcile performanceData Analysis & Business InsightsComparable metrics, slice analysis, uncertainty, failure clusters, and an independently checked ranking.
Live product

Turn experiment data into a real training monitor and inference experience, then prove it in-browser.

2 stages12 named roles

  1. Build live productFrontend InnovateA designed training monitor and inference experience backed by the experiment data contract.
  2. Prove live updatesBrowser Research & AcceptanceReal-page evidence that metrics refresh, inference works, responsive states hold, and console failures are absent.
Research & publication

Explain the winning design, render its evidence, position related work, write the paper, and audit it independently.

6 stages28 named roles

  1. Explain the designScientific Research DesignResearch questions, hypotheses, ablations, causal limits, and a defensible architecture narrative.
  2. Select the winning systemProduct ManagementOne decision register tying the chosen architecture to user value, evidence, trade-offs, and non-goals.
  3. Render publication figuresOffice DeliveryArchitecture, experiment, and comparison figures generated from the accepted source data with validation receipts.
  4. Map adjacent workPatent Landscape and Prior ArtA dated, query-reproducible adjacent-work landscape separating publications, claims, and open gaps.
  5. Draft the ACL paperResearch StudioA four-plus-page short paper whose method, results, figures, limitations, and citations share one evidence base.
  6. Audit the manuscriptAcademic Paper ReviewIndependent novelty, method, fact, citation, hallucination, organization, and concision findings with resolutions.
Reproduction & release

Reproduce the complete system, harden the publication boundary, and push one organized Mission repository.

3 stages12 named roles

  1. Reproduce & debugReview & DebugA clean-room reproduction of training, evaluation, inference, and site startup with root-cause fixes.
  2. Harden the repositoryCybersecurity AssuranceSecret, dependency, provenance, licence, workflow, and release-boundary assurance before publication.
  3. Publish the Mission repoBaseAn organized Git repository with exact setup, CUDA reproduction, data/model cards, dashboard, paper, evidence, and reviewed GitHub push.