Author as code
A Python questionnaire DSL with question types, branching, skip routing, randomization, quotas and piping.
Your study in
one repository.
Write the questionnaire, deploy a survey and run your analysis in a browser IDE. Keep the entire study together, versioned in Git.
Start freeOpen the live demoOpen beta · 30 days of Pro, free · No credit card
From question to report,
in the same project.
Wire the study together in siamang.yaml, then commit
and deploy without leaving the browser.
-
1
Author
Write the questionnaire as Python.
survey/questionnaire.py -
2
Validate
Every push is checked automatically.
Commit status -
3
Deploy
Publish a live survey URL.
siamang.yaml -
4
Collect
Responses land in a managed database.
PostgreSQL -
5
Analyze
Run scripts, build tables and ship reports.
analysis/ · reports/
A complete research
workspace in code.
Author, deploy, collect and analyze with the tools connected from the start.
Version every change
An in-browser editor, a full Git repository and commit history for every project.
Validate before fieldwork
Every push receives a valid, warning or error status. Catch broken logic before respondents see it.
Deploy in one click
Live survey URLs, streamed build logs and separate pilot and main environments.
Keep a managed database
A PostgreSQL schema per project. Browse responses and export CSV, Excel, SPSS, Stata, Parquet or SQLite.
Run reproducible analysis
A sandboxed Python SDK with direct data access for tables, weighting and reports.
Follow results live
Server-side, SQL-aggregated frequencies, crosstabs and respondent summaries.
Automate repeat work
Cron schedules, signed webhooks and connectors to Sheets, Supabase, S3 and more.
Security & governance
Keep control of
Keep control of
your research.
Source-available engine, per-response erasure and clean exports.
In beta: HIPAA and SOC 2 certification are not yet available.
Separate project data
Every project has its own PostgreSQL schema for responses.
Sandboxed execution
Validation and analysis run in ephemeral containers, with no network by default.
Encrypted secrets
Project secrets are encrypted at rest and write-only through the API.
Cloud plans
per organization.
Separate from Studio billing. Every new account gets 30 days of Pro free, then can subscribe or stay on Free. Your surveys and data are kept.
Scroll sideways to compare every plan
USD per organization, per month. A paid Cloud plan includes a commercial engine license for the code generated from your projects.
Cloud Pro, 50% off your first year.
Pre-order for $1,200 instead of $2,400. Billed once, activating at the 1.0 release planned for November 2026.
Cloud questions.
Can I export my data and leave?
Always. Export to CSV, Excel, SPSS .sav, Stata .dta, Parquet or SQLite at any time, or send responses straight to your own warehouse, bucket or database with a connector. The engine is source-available, so there is no lock-in.
What happens when I hit a limit?
Limits are enforced server-side with a clear upgrade prompt. Nothing is silently dropped. You stay in control of when to move up a tier.
Two ways to work.
The same engine.
Choose Cloud for Python and Git, or Studio for visual survey design and analysis. Accounts and billing are separate.
Siamang Cloud
Write the questionnaire and analysis as Python, in a Git repository with commit checks.
Start with CloudSiamang Studio
Build with clicks. Studio generates the Python for you. Open beta with 30 days of Pro free.
Explore StudioLearn it end to end.
A complete textbook and documentation for the engine and Cloud, from your first questionnaire to analysis and reporting.
Help shape Cloud 1.0.
The full research loop is working. Capacity is capped during beta and some Pro features are still rolling out. Version 1.0 is planned for November 2026.
Hear when Cloud 1.0 ships.
Leave your email for one launch notification.
Make your next study
reproducible.
Start with the full workflow and 30 days of Pro, free.
Start with Cloud
Open beta · 30 days of Pro, free