Evaluating single-turn agent interactions follows a sample that the majority groups perceive effectively. You present an enter, gather the output, and decide the end result. Frameworks like Strands Analysis SDK make this course of systematic by evaluators that assess helpfulness, faithfulness, and software utilization. In a earlier weblog submit, we coated find out how to construct complete analysis suites for AI brokers utilizing these capabilities. Nevertheless, manufacturing conversations not often cease at one flip.
Actual customers interact in exchanges that unfold over a number of turns. They ask follow-up questions when solutions are incomplete, change path when new info surfaces, and categorical frustration when their wants go unmet. A journey assistant that handles “Ebook me a flight to Paris” effectively in isolation may battle when the identical person follows up with “Truly, can we take a look at trains as a substitute?” or “What about resorts close to the Eiffel Tower?” Testing these dynamic patterns requires greater than static check circumstances with mounted inputs and anticipated outputs.
The core issue is scale as a result of you possibly can’t manually conduct a whole bunch of multi-turn conversations each time your agent modifications, and writing scripted dialog flows locks you into predetermined paths that miss how actual customers behave. What analysis groups want is a method to generate reasonable, goal-driven customers programmatically and allow them to converse naturally with an agent throughout a number of turns. On this submit, we discover how ActorSimulator in Strands Evaluations SDK addresses this problem with structured person simulation that integrates into your analysis pipeline.
Why multi-turn analysis is essentially more durable
Single-turn analysis has a simple construction. The enter is understood forward of time, the output is self-contained, and the analysis context is proscribed to that single trade. Multi-turn conversations break each considered one of these assumptions.
In a multi-turn interplay, every message is determined by all the things that got here earlier than it. The person’s second query is formed by how the agent answered the primary. A partial reply attracts a follow-up about no matter was unnoticed, a misunderstanding leads the person to restate their unique request, and a shocking suggestion can ship the dialog in a brand new path.
These adaptive behaviors create dialog paths that may’t be predicted at test-design time. A static dataset of I/O pairs, regardless of how giant, can’t seize this dynamic high quality as a result of the “right” subsequent person message is determined by what the agent simply mentioned.
Guide testing covers this hole in principle however fails in observe. Testers can conduct reasonable multi-turn conversations, however doing so for each state of affairs, throughout each persona sort, after each agent change is just not sustainable. Because the agent’s capabilities develop, the variety of dialog paths grows combinatorially, effectively past what groups can discover manually.
Some groups flip to immediate engineering as a shortcut, asking a big language mannequin (LLM) to “act like a person” throughout testing. With out structured persona definitions and express purpose monitoring, these approaches produce inconsistent outcomes. The simulated person’s conduct drifts between runs, making it troublesome to match evaluations over time or establish real regressions versus random variation. A structured method to person simulation can bridge this hole by combining the realism of human dialog with the repeatability and scale of automated testing.
What makes an excellent simulated person
Simulation-based testing is effectively established in different engineering disciplines. Flight simulators check pilot responses to situations that may be harmful or unattainable to breed in the actual world. Recreation engines use AI-driven brokers to discover thousands and thousands of participant conduct paths earlier than launch. The identical precept applies to conversational AI. You create a managed atmosphere the place reasonable actors work together along with your system below circumstances you outline, then measure the outcomes.
For AI agent analysis, a helpful simulated person begins with a constant persona. One which behaves like a technical professional in a single flip and a confused novice within the subsequent produces unreliable analysis information. Consistency means to keep up the identical communication model, experience degree, and persona traits by each trade, simply as an actual particular person would.
Equally vital is goal-driven conduct. Actual customers come to an agent with one thing they wish to accomplish. They persist till they obtain it, modify their method when one thing is just not working, and acknowledge when their purpose has been met. With out express objectives, a simulated person tends to both finish conversations too early or proceed asking questions indefinitely, neither of which displays actual utilization.
The simulated person should additionally reply adaptively to what the agent says, not observe a predetermined script. When the agent asks a clarifying query, the actor ought to reply it in character. If the response is incomplete, the actor follows up on no matter was unnoticed reasonably than shifting on. If the dialog drifts off subject, the actor steers it again towards the unique purpose. These adaptive behaviors make simulated conversations priceless as analysis information as a result of they train the identical dialog dynamics your agent faces in manufacturing.
Constructing persona consistency, purpose monitoring, and adaptive conduct right into a simulation framework is what differentiates structured person simulation from ad-hoc prompting. ActorSimulator in Strands Evals is designed round precisely these rules.
How ActorSimulator works
ActorSimulator implements these simulation qualities by a system that wraps a Strands Agent configured to behave as a practical person persona. The method begins with profile technology. Given a check case containing an enter question and an elective process description, ActorSimulator makes use of an LLM to create an entire actor profile. A check case with enter “I need assistance reserving a flight to Paris” and process description “Full flight reserving below funds” may produce a budget-conscious traveler with beginner-level expertise and an informal communication model. Profile technology provides every simulated dialog a definite, constant character.
With the profile established, the simulator manages the dialog flip by flip. It maintains the total dialog historical past and generates every response in context, protecting the simulated person’s conduct aligned with their profile and objectives all through. When your agent addresses solely a part of the request, the simulated person naturally follows up on the gaps. A clarifying query out of your agent will get a response that stays per the persona. The dialog feels natural as a result of each response displays each the actor’s persona and all the things mentioned up to now.
Purpose monitoring runs alongside the dialog. ActorSimulator features a built-in purpose completion evaluation software that the simulated person can invoke to guage whether or not their unique goal has been met. When the purpose is happy or the simulated person determines that the agent can’t full their request, the simulator emits a cease sign and the dialog ends. If the utmost flip depend is reached earlier than the purpose is met, the dialog additionally stops. This offers you a sign that the agent won’t be resolving person wants effectively. This mechanism makes certain conversations have a pure endpoint reasonably than working indefinitely or reducing off arbitrarily.
Every response from the simulated person additionally consists of structured reasoning alongside the message textual content. You’ll be able to examine why the simulated person selected to say what they mentioned, whether or not they had been following up on lacking info, expressing confusion, or redirecting the dialog. This transparency is effective throughout analysis growth as a result of you possibly can see the reasoning behind every flip, making it extra simple to hint the place conversations succeed or go off monitor.
Getting began with ActorSimulator
To get began, you will want to put in the Strands Analysis SDK utilizing: pip set up strands-agents-evals. For a step-by-step setup, you possibly can discuss with our documentation or our earlier weblog for extra particulars. Placing these ideas into observe requires minimal code. You outline a check case with an enter question and a process description that captures the person’s purpose. ActorSimulator handles profile technology, dialog administration, and purpose monitoring mechanically.
The next instance evaluates a journey assistant agent by a multi-turn simulated dialog.
from strands import Agent
from strands_evals import ActorSimulator, Case, Experiment
# Outline your check case
case = Case(
enter="I wish to plan a visit to Tokyo with resort and actions",
metadata={"task_description": "Full journey package deal organized"}
)
# Create the agent you wish to consider
agent = Agent(
system_prompt="You're a useful journey assistant.",
callback_handler=None
)
# Create person simulator from check case
user_sim = ActorSimulator.from_case_for_user_simulator(
case=case,
max_turns=5
)
# Run the multi-turn dialog
user_message = case.enter
conversation_history = []
whereas user_sim.has_next():
# Agent responds to person
agent_response = agent(user_message)
agent_message = str(agent_response)
conversation_history.append({
"position": "assistant",
"content material": agent_message
})
# Simulator generates subsequent person message
user_result = user_sim.act(agent_message)
user_message = str(user_result.structured_output.message)
conversation_history.append({
"position": "person",
"content material": user_message
})
print(f"Dialog accomplished in {len(conversation_history) // 2} turns")The dialog loop continues till has_next() returns False, which occurs when the simulated person’s objectives are met or simulated person determines that the agent can’t full the request or the utmost flip restrict is reached. The ensuing conversation_history incorporates the total multi-turn transcript, prepared for analysis.
Integration with analysis pipelines
A standalone dialog loop is helpful for fast experiments, however manufacturing analysis requires capturing traces and feeding them into your evaluator pipeline. The subsequent instance combines ActorSimulator with OpenTelemetry telemetry assortment and Strands Evals session mapping. The duty perform runs a simulated dialog and collects spans from every flip, then maps them right into a structured session for analysis.
from opentelemetry.sdk.hint.export import BatchSpanProcessor
from opentelemetry.sdk.hint.export.in_memory_span_exporter import InMemorySpanExporter
from strands import Agent
from strands_evals import ActorSimulator, Case, Experiment
from strands_evals.evaluators import HelpfulnessEvaluator
from strands_evals.telemetry import StrandsEvalsTelemetry
from strands_evals.mappers import StrandsInMemorySessionMapper
# Setup telemetry for capturing agent traces
telemetry = StrandsEvalsTelemetry()
memory_exporter = InMemorySpanExporter()
span_processor = BatchSpanProcessor(memory_exporter)
telemetry.tracer_provider.add_span_processor(span_processor)
def evaluation_task(case: Case) -> dict:
# Create simulator
user_sim = ActorSimulator.from_case_for_user_simulator(
case=case,
max_turns=3
)
# Create agent
agent = Agent(
system_prompt="You're a useful journey assistant.",
callback_handler=None
)
# Accumulate spans throughout dialog
all_target_spans = []
user_message = case.enter
whereas user_sim.has_next():
memory_exporter.clear()
agent_response = agent(user_message)
agent_message = str(agent_response)
# Seize telemetry
turn_spans = checklist(memory_exporter.get_finished_spans())
all_target_spans.prolong(turn_spans)
# Generate subsequent person message
user_result = user_sim.act(agent_message)
user_message = str(user_result.structured_output.message)
# Map to session for analysis
mapper = StrandsInMemorySessionMapper()
session = mapper.map_to_session(
all_target_spans,
session_id="test-session"
)
return {"output": agent_message, "trajectory": session}
# Create analysis dataset
test_cases = [
Case(
name="booking-simple",
input="I need to book a flight to Paris next week",
metadata={
"category": "booking",
"task_description": "Flight booking confirmed"
}
)
]
evaluator = HelpfulnessEvaluator()
dataset = Experiment(circumstances=test_cases, evaluator=evaluator)
# Run evaluations
report = Experiment.run_evaluations(evaluation_task)
report.run_display()
This method captures full traces of your agent’s conduct throughout dialog turns. The spans embrace software calls, mannequin invocations, and timing info for each flip within the simulated dialog. By mapping these spans right into a structured session, you make the total multi-turn interplay obtainable to evaluators like GoalSuccessRateEvaluator and HelpfulnessEvaluator, which might then assess the dialog as an entire, reasonably than remoted turns.
Customized actor profiles for focused testing
Automated profile technology covers most analysis situations effectively, however some testing objectives require particular personas. You may wish to confirm that your agent handles an impatient professional person in another way from a affected person newbie, or that it responds appropriately to a person with domain-specific wants. For these circumstances, ActorSimulator accepts a totally outlined actor profile that you just management.
from strands_evals.varieties.simulation import ActorProfile
from strands_evals import ActorSimulator
from strands_evals.simulation.prompt_templates.actor_system_prompt import (
DEFAULT_USER_SIMULATOR_PROMPT_TEMPLATE
)
# Outline a customized actor profile
actor_profile = ActorProfile(
traits={
"persona": "analytical and detail-oriented",
"communication_style": "direct and technical",
"expertise_level": "professional",
"patience_level": "low"
},
context="Skilled enterprise traveler with elite standing who values effectivity",
actor_goal="Ebook enterprise class flight with particular seat preferences and lounge entry"
)
# Initialize simulator with customized profile
user_sim = ActorSimulator(
actor_profile=actor_profile,
initial_query="I must ebook a enterprise class flight to London subsequent Tuesday",
system_prompt_template=DEFAULT_USER_SIMULATOR_PROMPT_TEMPLATE,
max_turns=10
)
By defining traits like endurance degree, communication model, and experience, you possibly can systematically check how your agent performs throughout completely different person segments. An agent that scores effectively with affected person, non-technical customers however poorly with impatient consultants reveals a particular high quality hole you could tackle. Working the identical purpose throughout a number of persona configurations turns person simulation right into a software for understanding your agent’s strengths and weaknesses by person sort.
Greatest practices for simulation-based analysis
These greatest practices enable you get essentially the most out of simulation-based analysis:
- Set
max_turnsprimarily based on process complexity, utilizing 3-5 for centered duties and 8-10 for multi-step workflows. If most conversations attain the restrict with out finishing the purpose, enhance it. - Write particular process descriptions that the simulator can consider in opposition to. “Assist the person ebook a flight” is simply too obscure to guage completion reliably, whereas “flight reserving confirmed with dates, vacation spot, and worth” provides a concrete goal.
- Use auto-generated profiles for broad protection throughout person varieties and customized profiles to breed particular patterns out of your manufacturing logs, resembling an impatient professional or a first-time person.
- Deal with patterns throughout your check suite reasonably than particular person transcripts. Constant redirects from the simulated person means that the agent is drifting off subject, and declining purpose completion charges after an agent change factors to a regression.
- Begin with a small set of check circumstances overlaying your most typical situations and broaden to edge circumstances and extra personas as your analysis observe matures.
Conclusion
We confirmed how ActorSimulator in Strands Evals permits systematic, multi-turn analysis of conversational AI brokers by reasonable person simulation. Quite than counting on static check circumstances that seize solely single exchanges, you possibly can outline objectives and personas and let simulated customers work together along with your agent throughout pure, adaptive conversations. The ensuing transcripts feed straight into the identical analysis pipeline that you just use for single-turn testing, providing you with helpfulness scores, purpose success charges, and detailed traces throughout each dialog flip.
To get began, discover the working examples within the Strands Brokers samples repository. For groups evaluating brokers deployed by Amazon Bedrock AgentCore, the next AgentCore evaluations pattern display find out how to simulate interactions with deployed brokers. Begin with a handful of check circumstances representing your most typical person situations, run them by ActorSimulator, and consider the outcomes. As your analysis observe matures, broaden to cowl extra personas, edge circumstances, and dialog patterns.
Concerning the authors
