market-pilot
Evidence-grounded market research prototype with traceable AI workflows.
What is it?
What it is
Market Pilot is an evidence-grounded market research prototype that converts public market signals into traceable research workflows, emphasizing source-grounded evidence and transparency.
Why it exists
To address the gap in AI-generated market summaries by providing a structured workflow that preserves source signals, separates evidence from inference, masks identifiers, and highlights uncertainty in reports.
Who should use it
Product managers, market researchers, AI developers, and open‑source contributors who want to prototype evidence‑grounded market research workflows using React and FastAPI.
Who should avoid it
Absolute beginners with no experience in JavaScript, Python, or web frameworks, or teams looking for a turnkey production system without customization.
How it works
A quick walkthrough in plain English
How market-pilot works
Step 1 of 3
You interact with it
Open market-pilot, send a request, or connect it to your stack.
Features
Advantages
- Evidence-grounded and traceable research workflows
- Privacy-preserving through de-identification and masking
- Bilingual support (English/Simplified Chinese) with localized reports
- Open-source Apache-2.0 license encourages community contributions
- Modular architecture (React frontend, FastAPI backend) enables extensibility
- Caching reduces redundant computation and speeds up report retrieval
- Clear separation of evidence and inference improves interpretability
- Supports switching between public and authenticated experiences
Disadvantages
- Currently a prototype; some advanced workflows require user-provided credentials
- Demo data must be labeled as sanitized; real data connectors need authorized provider credentials
- Setup requires specific versions (Node.js 24+, npm 11+, Python 3.13+) which may be restrictive
- Technical knowledge needed to configure environment variables and run services
- Limited to processing public signals; private or proprietary data sources not supported out-of-the-box
- Localized report reading depends on pre-built cache; missing cache requires re-running research
Installation
native
git clone https://github.com/Dgeloe4-yb/market-pilot.git Set-Location market-pilot npm.cmd ci python -m venv .venv .\venv\Scripts\python.exe -m pip install -r collector\requirements.txt # review .env.example and export variables # start API Set-Location collector ..\.venv\Scripts\python.exe -m uvicorn api.app:app --port 8001 # start frontend Set-Location .. npm.cmd run dev
FAQ
What are the prerequisites for setting up Market Pilot locally?
You need Node.js 24+, npm 11+, Python 3.13+, and Git. After cloning the repo, run npm.cmd ci, create a Python virtual environment, and install the collector requirements from collector/requirements.txt.
How do I start the backend API and frontend development servers?
From the collector directory, run ..\.venv\Scripts\python.exe -m uvicorn api.app:app --port 8001 to start the FastAPI backend on port 8001. In another terminal at the repository root, run npm.cmd run dev to start the Vite frontend, which will be available at http://localhost:5173.
Do I need to provide my own API credentials to use Market Pilot, and where should I set them?
Yes. The demo uses sanitized data; any real data collection or model-backed runs require your own authorized provider credentials. Review .env.example, then export the needed variables into your process environment (the app does not auto-load .env files).
How does Market Pilot handle localization and language switching?
The product supports English and Simplified Chinese. Public URLs are unprefixed for English and prefixed with /zh for Chinese. Locale is resolved account-aware, and report localizations are cached so you can read them without re‑collecting data or re‑running research.
What data and privacy guidelines should I follow when contributing or running Market Pilot?
Do not commit credentials, raw scraped data, user images, local databases, or production run outputs. Use only data you are authorized to collect and process. Public examples must be synthetic or explicitly sanitized. See the Data and privacy section in the README for more details.
Where can I find information about the project's architecture, contributing guidelines, and license?
The architecture is detailed in docs/architecture.md. Contributing guidelines are in CONTRIBUTING.md and security reports in SECURITY.md. The project is licensed under Apache-2.0; see the LICENSE file.
Featured in Videos
YouTube tutorials and walkthroughs for market-pilot
Alternatives
Similar projects ranked by category, topics, and text overlap.
