Digital Twins
Real target audience feedback in seconds. Based on 1M+ real profiles.
ONE KEY · 7 DOMAINS · 88 ENDPOINTS
One API for the entire neuroflash stack.
Query a million real people, create on-brand content and images, straight from your code. Digital Twins, Content, Brand Voice and Images behind a single key.
Includes around 5,000 free credits per month, self-serve, no subscription.
# Query a real audience segment from your code curl https://api.neuroflash.com/v1/twins/query \ -H "Authorization: Bearer $NF_KEY" \ -d '{ "segment": "gen_z_de", "prompt": "Rate this slogan 1-10 and say why.", "input": "Build it like you mean it." }' # response { "segment": "gen_z_de", "score": 7.4, "rationale": "Direct and confident, but...", "profiles_polled": 240, "credits": 18 }
PLATFORM
Everything neuroflash can do, callable from your code.
88 endpoints across 7 domains. Use them standalone or chain them into a workflow the app would never allow.
Digital Twins
Query 1M+ real profiles. One twin, a group or a 500-person panel.
Content
Generate copy for any channel, in your brand voice, at scale.
Brand Voice
Create, import and apply voices, from a URL or text sample.
Images
On-brand image generation, straight into your pipeline.
Audiences
List segments, inspect data points, build your own audiences.
Workspaces
Accounts, keys, environments and usage, managed via API.
Insights
Compare results across segments and runs programmatically.
See for yourself how Digital Twins respond.
Pick a use case and an audience, and see how differently the twins react.
Question to the twins
Two ads for a software subscription: A: “Save 30% on your first subscription” — B: “Try it free for 14 days, no credit card”. Which are you more likely to click?
Switch audience
DIGITAL TWIN
Pre-produced Digital Twins responses · Based on 1M+ real survey profiles
HOW TO START
Connected in about a minute.
Paste one URL, sign in via OAuth, done. No code, no installation.
01
Open your tool
Claude Desktop, Cursor, Windsurf or another MCP tool. There is nothing to install, the neuroflash MCP server runs in the cloud.
02
Add the connector
Settings, connectors, paste the server URL and sign in once via OAuth with your neuroflash account.
03
Just ask
“What does Gen Z think of our slogan?” The Digital Twins answer right in your chat.
Works with Claude Desktop
Cursor
Windsurf
and any MCP client via HTTP streaming
WORKFLOW
From key to production in three calls.
Authenticate
Create a key
Create a service account in the neuroflash settings and generate a key. Works without an app subscription.
export NF_KEY=nf_live_…
Call
Send a request
One POST: choose a segment and send your prompt.
POST /v1/twins/query
Ship
Integrate
Read the JSON into your app, test and scale.
200 OK · 18 Credits
COMBINATION
The real value is in the combination.
One workflow, three domains: write in your brand voice, test with a real segment, then revise based on the score. The app runs this as a fixed path, the API lets you build your own.
// 1. generate in your brand voice const draft = await nf.content.generate({ voiceId, brief }); // 2. test against a real audience segment const test = await nf.twins.query({ segment: "b2b_dm_de", input: draft.text }); // 3. branch on the score if (test.score < 7) return nf.content.revise({ draft, feedback: test.rationale });
PRICING
Its own pricing, separate from your app quota.
One unit, API credits, powers both API and MCP, independent of your app subscription. Start free, scale with usage.
BUILT FOR PRODUCTION
Built for production.
The reliability and security checks a developer runs before committing.
OAuth 2.0 + PKCE
Servers in Germany
ISO 27001 certified
99.9% SLA from Scale
status.neuroflash.com
Versioned, with a public changelog
Rate limits per tier, published
OUR DIGITAL TWINS VS LLM PERSONAS
Why not just use ChatGPT?
neuroflash Digital Twins are based on real survey data.
LLM personas are based on assumptions.
Data foundation
Output
Grounding
neuroflash Digital Twins
1M+ real survey profiles since 2017
Real individual personality
Human, in real data
Generic LLM personas
Prompt-based, no real data foundation
Average generalization
Invented in the prompt
Data foundation
neuroflash Digital Twins
1M+ real survey profiles since 2017
Generic LLM personas
Prompt-based, no real data foundation
Output
neuroflash Digital Twins
Real individual personality
Generic LLM personas
Average generalization
Grounding
neuroflash Digital Twins
Human, in real data
Generic LLM personas
Invented in the prompt
FAQ
The questions a developer actually has.
How do I authenticate?
Create a service account in your neuroflash settings and send your key as a bearer credit, with OAuth 2.0 + PKCE for delegated access. You don’t need an app subscription to get a key.
What are the rate limits?
60 requests/min on Developer, 600 on Scale (can be raised), custom on Enterprise. Limits are published, not hidden behind sales.
What is the SLA, and where is the status page?
Best-effort on Developer, 99.9% from Scale. Live status at status.neuroflash.com.
What happens to my data?
They’re only processed to handle your request, on servers in Germany, to ISO 27001, and your inputs never train public models.
READY?
Stop guessing. Start predicting.
Get a free key and build on a million real people. Your first call takes less than five minutes, no subscription, no sales call.
