Your AI-powered companion for mastering Palantir Foundry
238 flashcards • 310 quiz questions • 12 hands-on labs • 51 platform mappings • 3 interview modes
Getting Started • Features • Screenshots • Architecture • API Reference
Ontologist is a full-stack study platform for professionals preparing to work with Palantir Foundry. It provides a structured 14-section curriculum that takes you from platform foundations through Python/PySpark, data science, application building, AIP integration, and FDE soft skills — with AI-powered tutoring at every step.
The app personalizes the entire experience to your background. Upload your resume, tell it what platforms you know, and every AI interaction — from the tutor to mock interviews to lab reviews — adapts to speak your language.
 |
 |
| Personalized onboarding wizard | Ongoing profile management |
# Clone and install
git clone <repo-url> && cd ontologist
npm install && cd client && npm install && cd ..
# Configure
cp .env.example .env
# Add your Anthropic API key to .env
# Launch
npm run devOpen http://localhost:5173 — you'll be greeted by the onboarding wizard on first run.
docker run -d \
-p 3001:3001 \
-e ANTHROPIC_API_KEY=your-key-here \
-v ontologist-data:/app/data \
ghcr.io/windoze95/ontologist:latestOpen http://localhost:3001 after the container starts.
-
Copy the template to your Unraid flash drive:
/boot/config/plugins/dockerMan/templates-user/my-ontologist.xmlDownload it from
unraid/ontologist.xmlin this repo. -
In the Unraid web UI, go to Docker tab → Add Container → select Ontologist from the template dropdown.
-
Enter your Anthropic API Key and click Apply.
The container uses br0 networking and gets its own IP. The web UI will be available at http://<container-ip>. Data is persisted to /mnt/user/appdata/ontologist by default.
A structured path from zero to Foundry-proficient, organized into focused sections with flashcards, quizzes, labs, and reference materials.
 |
 |
| Section detail with activities and resources | Rosetta Stone: map what you know to Foundry |
| Section | Topic | Highlights |
|---|---|---|
| 1 | Foundry Foundations & Platform Architecture | Security model, navigation, tech stack |
| 2 | Datasets, Branches & Data Connections | Connectors, branching, agents, Media Sets |
| 3 | The Ontology Deep Dive | Object Types, Links, Actions, Functions, OMS/OSS |
| 4 | Pipeline Builder & Data Expectations | Visual pipelines, scheduling, monitoring |
| 5 | Code Repositories — Python & PySpark | @transform, DataFrame ops, joins, window functions |
| 6 | Advanced Transforms & Incremental Pipelines | @incremental, performance, schema evolution |
| 7 | Code Workbooks & pandas in Foundry | @pandas_transform, SQL/R cells, visualization |
| 8 | Data Science & ML in Foundry | scikit-learn, PyTorch, model deployment, ML Objectives |
| 9 | Workshop — Operational Applications | Widgets, Variables, Events, Modules, Actions |
| 10 | Slate, Contour & Quiver | Custom apps, dashboards, exploration tools |
| 11 | Functions, OSDK & External Integration | TypeScript Functions, REST API, webhooks |
| 12 | AIP — AI-Powered Foundry | AIP Logic, AIP Assist, LLM integration, RAG |
| 13 | FDE Skills — Decomposition & Communication | Stakeholder discovery, engagement management |
| 14 | Full Simulation & Comprehensive Assessment | End-to-end engagement sim, 50-question final |
Spaced repetition using Leitner boxes (1-5). Cards advance on correct answers, retreat on incorrect. Each card includes a platform analogy to bridge your existing knowledge.
238 cards across 13 categories: platform, data, ontology, pipeline, pyspark, transforms, workbooks, ml, workshop, analytics, integration, aip, fde-skills.
 |
 |
| Desktop flashcard review | Fully responsive on mobile |
Timed multiple-choice with immediate feedback and explanations. Technical sections include PySpark, pandas, TypeScript, and OSDK code snippets.
310 questions — 20 per section plus a 50-question comprehensive final assessment.
Interactive two-column comparison mapping 51 concepts from platforms you already know to their Foundry equivalents. Organized across 9 categories with expandable detail rows.
12 multi-step labs (60 steps total) with progressive hints and AI-powered work review. Labs build on a continuous scenario thread personalized to your company context.
 |
 |
| Step-by-step guided walkthrough | Full engagement simulation with your company |
AI-powered practice across three Palantir interview formats. Includes timer, transcript saving, and AI-generated scorecards.
| Mode | Duration | Focus |
|---|---|---|
| Decomposition | 45 min | Breaking down complex, ambiguous business problems |
| Behavioral | 30 min | STAR-format questions with Palantir cultural emphasis |
| Technical | 30 min | Data modeling, system design, practical problem-solving |
Dynamically generates exercises from your profile accomplishments. Practice reframing your real experience in Foundry terminology, with AI feedback on each translation.
Track overall completion, section-by-section breakdown, flashcard mastery distribution, quiz scores, and time spent.
A collapsible chat panel on every page. Supports multiple AI modes (Tutor, Interviewer, Lab Reviewer, Feedback) and renders full markdown responses. Every prompt is personalized to your background.
The entire app works on mobile devices with an adaptive layout — collapsible sidebar, touch-friendly cards, and responsive grids.
| Layer | Technology |
|---|---|
| Frontend | React 19 + Vite 7 + Tailwind CSS v4 |
| Backend | Node.js + Express 5 |
| Database | SQLite (better-sqlite3) |
| AI | Anthropic Claude API (claude-sonnet-4-6) |
| Icons | lucide-react |
| Routing | react-router-dom v7 |
ontologist/
├── client/ # React frontend (Vite)
│ └── src/
│ ├── components/
│ │ ├── Layout/ # Sidebar, TopBar, AIChatPanel
│ │ ├── StudyPlan/ # Dashboard + SectionDetail
│ │ ├── Flashcards/ # Spaced repetition engine
│ │ ├── Quiz/ # Timed quiz with scoring
│ │ ├── RosettaStone/ # Platform comparison viewer
│ │ ├── Labs/ # Multi-step guided walkthroughs
│ │ ├── MockInterview/ # AI interview simulator
│ │ ├── TranslateExperience/ # Experience reframing exercises
│ │ ├── Onboarding/ # 5-step profile setup wizard
│ │ ├── Profile/ # Profile management page
│ │ ├── Guards/ # Route guard (RequireProfile)
│ │ └── Progress/ # Analytics dashboard
│ ├── data/ # Static content (cards, quizzes, labs, mappings)
│ ├── hooks/ # useAI, useProgress, useTimer
│ └── context/ # AppContext (global state)
├── server/
│ ├── routes/
│ │ ├── ai.js # Claude API proxy + SSE streaming
│ │ ├── progress.js # Progress tracking CRUD
│ │ └── profile.js # Profile CRUD + resume parsing
│ ├── utils/
│ │ └── promptTemplate.js # Dynamic prompt injection from profile
│ ├── db/
│ │ ├── schema.sql # SQLite schema
│ │ └── seed.js # DB initialization
│ └── prompts/ # 8 AI system prompt templates
└── data/ # SQLite database (auto-created)
Four tables in SQLite (data/progress.db), auto-created on first run:
| Table | Purpose |
|---|---|
| user_profile | Name, company, background, accomplishments, resume text |
| progress | Activity completion tracking per section |
| flashcard_progress | Leitner box state per card (box 1-5, review counts) |
| quiz_results | Quiz answers and scores per attempt |
| interview_transcripts | Full chat transcripts with AI feedback |
| Method | Path | Description |
|---|---|---|
POST |
/api/ai/chat |
Send messages to Claude (non-streaming) |
POST |
/api/ai/chat/stream |
Send messages to Claude (SSE streaming) |
Request: { messages: [{role, content}], mode: string }
Modes: tutor interviewer-decomp interviewer-behavioral interviewer-technical feedback lab-reviewer decomposition resume-refiner
| Method | Path | Description |
|---|---|---|
GET |
/api/profile |
Get user profile |
PUT |
/api/profile |
Create or update profile |
POST |
/api/profile/resume |
Upload and extract resume text (PDF/TXT) |
POST |
/api/profile/resume/parse |
AI-parse resume into structured fields |
GET |
/api/profile/exists |
Lightweight existence check |
| Method | Path | Description |
|---|---|---|
GET |
/api/progress |
All progress entries |
GET |
/api/progress/section/:section |
Progress for a section |
POST |
/api/progress |
Create/update progress entry |
GET |
/api/progress/flashcards |
All flashcard review state |
POST |
/api/progress/flashcards |
Record a flashcard review |
GET |
/api/progress/quizzes |
All quiz results |
POST |
/api/progress/quizzes |
Save a quiz result |
GET |
/api/progress/interviews |
All interview transcripts |
POST |
/api/progress/interviews |
Save an interview transcript |
GET |
/api/progress/stats |
Aggregate statistics |
| Command | Description |
|---|---|
npm run dev |
Start frontend + backend concurrently |
npm run dev:server |
Backend only (nodemon) |
npm run dev:client |
Vite dev server only |
npm run build |
Production build |
npm start |
Production server |
npm run seed |
Initialize database |
npm run reset |
Delete and re-initialize database |
| Variable | Description | Default |
|---|---|---|
ANTHROPIC_API_KEY |
Anthropic API key | (required for AI features) |
PORT |
Backend server port | 3001 |
DATABASE_PATH |
SQLite database path | ./data/progress.db |
All content — flashcards, quizzes, Rosetta Stone mappings, labs, and the study plan — is bundled into the frontend. The app works fully offline for everything except AI-powered features (tutor, interviews, lab review, resume parsing).
- Theme: Dark mode with navy/slate backgrounds, electric blue accents
- Typography: Inter (body) + JetBrains Mono (code)
- Components: Rounded-xl cards, navy-600 borders, 200ms transitions
- Colors: Defined via Tailwind v4
@themeinclient/src/index.css
Built with React, Express, SQLite, and Claude