Division 01 — Synthetic Persona Systems

We simulate people
so you don't test on real ones.

Before a drug reaches patients, it's tested on a model. Before a car hits the road, it hits a dummy. Chatbots, ads, and AI products still go straight to real people. We build the dummies — and the crash-test rigs.

Test 02 — Audiences API · Persona Library

Personas you can query,
not paragraphs pasted into
a system prompt.

Every persona is a testable hypothesis — trait profiles validated through cross-model pipelines, not demographic cutouts. We maintain a library spanning cognitive styles, personality structures, cultural contexts, and vulnerability profiles.

Quality depends on signal-bearing features, not descriptor volume. Adding more persona details does not improve output coherence. We select the traits that matter and validate them through independent generate, simulate, and evaluate pipelines.

Field deployments
Spinka Film Studio Fobos BizYou
01
Static Personas
Query-ready, reproducible profiles for surveys, message testing, and concept validation. Ask the same question to 200 personas, get a distribution. RESTful endpoints, trait-vector selection, batch runs.
APIBatchAvailable
02
Dynamic Personas
Evolving conversational partners for stress-testing chatbots. Mood shifts, trust erodes, patience runs out. Designed for multi-turn evaluation where single-reply testing is meaningless.
Multi-turnAPIAvailable
03
Curated Chat Gallery
A rotating selection of notable personas available for live conversation via web UI. Explore the range before you integrate.
Web UILive
04
Custom Persona Commissioning
Need a specific profile we don't have? We build and validate it against your spec, delivered as an API-accessible persona — static or dynamic.
Bespoke2–4 weeks
Test 03 — Agents Async Systems · Email + Web

AI systems that do the
thinking, not the chatting.

Agents are complete analytical systems. They orchestrate entire research workflows end-to-end: assemble a persona panel, run structured interviews, collect responses, then hand the raw material to a separate expert agent that extracts patterns, flags contradictions, and delivers a structured insight report.

You can reach agents through a web interface or a dedicated email address. Send an email, get a report. Paste a draft, get a critique. No login, no API key, no onboarding deck.

Field deployments
myreviewer.one Lawsome
01
Clarity
Our flagship decision-support agent. Upload a problem, get structured multi-perspective analysis with explicit uncertainty and dissent mapping.
Web AppAvailable
02
Panel Runs with Insight Extraction
Define a brief, select a persona panel, and let the agent run interviews or focus groups autonomously. A separate expert agent analyzes transcripts and delivers structured insights — pattern extraction with supporting evidence, not a summary.
OrchestratedAvailable
03
Email-Accessible Agents
Configure a dedicated inbox. Send your brief as an email, receive the agent's output as a reply. Built for teams that live in Outlook, not in dev tools.
EmailNo-login
04
Custom Agent Design
We build agents for specific workflows: message pre-testing, policy stress-testing, communication auditing, synthetic user research.
Bespoke
Test 04 — Auditors Pre-Prod Stress · Behavioural Audit

We crash-test your AI
before your users do.

You've built a chatbot, a companion, a customer-facing AI system. How do you know it behaves correctly under pressure? Not from your QA team's polite test prompts — from sustained, psychologically realistic interactions with users who push, escalate, misunderstand, and need help.

We run two auditing systems: BSB, our proprietary behavioural stress-testing battery, and Reverse Audiences, where we send our personas to talk to your system and deliver annotated transcripts and risk reports.

01
BSB Behavioural Stress Test
Pre-production battery for LLM-based systems. Hundreds of automated runs across edge cases, adversarial inputs, and multi-turn pressure sequences. Structured report with pass/fail thresholds and severity tiers. Think load testing, but for behaviour.
ProprietaryAvailable
02
Reverse Audiences Audit
Our personas initiate conversations with your AI system. We monitor for safety failures, inappropriate escalation, vulnerability exploitation, policy violations. Annotated turn-by-turn transcripts included.
AnnotatedAvailable
03
Continuous Monitoring
Recurring audits on a schedule. Catch drift, regression, and emergent failure modes before your users do.
Subscription
04
Trajectory Observatory
Our public experimental system tracking multi-turn conversations across models and prompt configurations at turn-level granularity. Not a sample report — a window into the annotation depth and analytical resolution we bring to every audit.
PublicOpen
Test 05 — Methodology Validation · Measurement

Built on experimental psychology,
not prompt engineering.

We don't generate personas by writing longer system prompts. We separate creation, simulation, and evaluation across independent models — eliminating preference leakage. We validate against national survey data. We run hundreds of iterations per condition and analyze distributions, not cherry-picked outputs.

Our finding: newer, more capable models are often more sensitive to minor perturbations. A name change shifts scores. A formatting tweak alters reasoning. Single-shot evaluation hides this. We don't do single-shot evaluation.

Test 06 — Application Case Files · Industry

Built for teams that need
more than a demo.

Case 01Product
AI Product Teams
Shipping a chatbot, companion, or assistant? Test it against realistic users before they find the edge cases.
AudiencesAuditors
Case 02Marketing
Marketing & Comms
Pre-test messaging, campaigns, and crisis comms against psychographically rich synthetic panels. No recruitment.
AudiencesAgents
Case 03Compliance
Compliance & Regulatory
Audit AI systems against safety, fairness, and policy requirements. Independent evaluation, structured evidence.
Auditors
Case 04Public Sector
Government
Stress-test policy language, public communications, and citizen-facing AI against diverse population segments.
AgentsAudiences
Case 05Social AI
Companion & Social AI
Your product simulates relationships. Ours simulates the vulnerable users yours will encounter. Red-team before they arrive.
Auditors
Case 06Research
Research & Academia
Replicate survey studies, test hypotheses on synthetic populations, validate your own methods against ours.
Audiences
Test 07 — Observatory Public · Live

75 annotated trajectories where
LLMs failed vulnerable users.

A public window into our annotation depth. Turn-by-turn, model-by-model, the Vulnerable Trajectory Observatory tracks where commercial AI systems lose the plot — so the industry can stop pretending the problem is somewhere else.

Test 08 — Contact Scoping · Pilot

Tell us what you're building.
We'll tell you what breaks.

Raw personas via API, an async agent for a specific workflow, or a behavioural audit of your production AI — start here. No discovery decks. No six-week procurement waltz.

hello@impersonato.com ↗