Skip to main content
The gap between an agent using tools and a deep research system is orchestration, grounding, and auditable deliverables. Agno lets you build a deterministic, step-based research pipeline in a few lines. Run steps in parallel, in loops, or one after the other. 
from agno.workflow import Parallel, Step, Workflow

investment_workflow = Workflow(
    id="investment-workflow",
    name="Investment Review Pipeline",
    steps=[
        Step(name="Market Assessment", agent=market_analyst),
        Parallel(
            Step(name="Fundamental Analysis", agent=financial_analyst),
            Step(name="Technical Analysis", agent=technical_analyst),
            name="Deep Dive",
        ),
        Step(name="Risk Assessment", agent=risk_officer),
        Step(name="Investment Memo", agent=memo_writer),
        Step(name="Committee Decision", agent=committee_chair),
    ],
)

result = investment_workflow.run("Run a full investment review on NVDA")
# result.content holds the Committee Decision. Same steps, every run.
That is an auditable five-step review with a parallel deep-dive. Swap the agents and the mandate and the same shape researches anything.

​
Learn How To

ConcernPage
Pick the orchestration shapeOrchestration patterns
Run specialists in parallelParallel investigation
Ground every agentGrounding research
End in an auditable artifactStructured deliverable
Make research compoundInstitutional learning
Put it behind a surfaceServe and embed

​
Explore

Orchestration patterns

Route, coordinate, broadcast, task, and deterministic pipelines.

Parallel investigation

Fan specialists out at once, synthesize the results.

Grounding research

Mandate in the prompt, a library in RAG, prior work on disk.

Structured deliverable

End in a memo and a decision, with citations and conviction.

Institutional learning

Shared learnings so the next review starts ahead.

​
Developer Resources