By codesPACT — Infrastructure & AI Division
Learning a new language often feels like juggling disconnected tools, static vocabulary lists, and repetitive exercises. At codesPACT, we believe learning should adapt to the user; not the other way around.
That’s why we created Ankineitor, an AI-powered system that turns any document into structured learning material in just a few clicks. Built by our Infrastructure & AI Lead, Aurelio Márquez, Ankineitor blends automation, NLP, and smart design to create a seamless document-to-learning pipeline.
Why Ankineitor? A Smarter Way to Learn From What You Already Read
Most language-learning platforms rely on predefined word lists and static content. Ankineitor flips the traditional model: instead of adjusting to the app’s curriculum, your content becomes the curriculum.
Whether you read articles, books, study materials, work documents, or academic papers, Ankineitor helps you learn directly from them—without friction, configuration overload, or manual filtering.
Upload a file. Extract vocabulary. Generate sentences. Build flashcards. All automated. All tailored to you.
Introducing the Word Extractor: A Complete Vocabulary Engine
One of the newest components of Ankineitor is the Word Extractor, designed to make vocabulary analysis faster, cleaner, and more efficient.
Here’s how it elevates the entire learning process:
1. Upload Any Supported File Format
You can upload:
- DOCX
- TXT
- PPTX
With one click on Analyze Text, Ankineitor processes the file and extracts its linguistic structure.
2. Identify the Most Frequent and Relevant Words
Using token-frequency analysis and NLP preprocessing, the system detects:
- High-frequency vocabulary
- Patterns across the entire document
- Key terms that matter most for your learning context
This transforms raw text into actionable data.
3. Remove HSK Levels 1–5 Instantly
Aurelio added a highly requested feature:
automatic filtering of HSK-level vocabulary.
If you’re learning Chinese, you can exclude common beginner terms (HSK 1–5), letting you focus exclusively on more advanced or unfamiliar words.
This feature alone saves hours of manual screening.
4. Apply Frequency Thresholds
You can set a minimum frequency filter to remove rare or irrelevant words.
This keeps your list clean, focused, and efficient for actual learning.
5. Generate a Final Word List, to be Ready for the Pipeline
Once filtered, Ankineitor provides a clean, ready-to-copy word list that you can send into the next steps:
- AI sentence generation
- Flashcard creation (CSV / HTML)
- Export for Anki or EdTech platforms
The entire workflow takes seconds.
From Words to Learning: The Full Ankineitor Pipeline
While the Word Extractor handles vocabulary analysis, Ankineitor continues the process with two powerful components.
AI Sentence Generator
Every word in the final list is processed through NLP models that create:
- Natural sentences
- Context-aware examples
- Language-consistent outputs
This step adds meaning, not just definitions.
Flashcard Builder
The system compiles everything into:
- CSV templates (compatible with Anki or spaced-repetition systems)
- HTML templates for web or mobile use
Your document becomes a fully structured learning deck.
What Makes Ankineitor Unique
- Multi-language support (EN, ES, CN, FR, IT)
- Optional translations
- HSK-level filtering
- Frequency-based vocabulary extraction
- AI sentence generation
- Scalable cloud infrastructure
- CSV/HTML card builder
- Built for EdTech integrations and personal learning workflows
It’s more than a demo — it’s a functional, configurable, scalable platform.
Built With codesPACT Engineering Standards
Ankineitor reflects the engineering philosophy of codesPACT:
- Clean, modular architecture
- Cloud-first deployment (GCP, Docker, Kubernetes)
- Automated CI/CD (GitHub Actions)
- Python-based NLP engine
- Data processing pipelines designed for speed and clarity
The system is lightweight, extendable, and ready to evolve.
See Ankineitor in Action
Visit the live demo:
👉 ankineitor.codespact.com
If you want to explore the GitHub repository, Behance showcase, or a technical deep dive article, we can prepare those as well.
Final Thoughts
Ankineitor is part of our commitment to creating intelligent tools that solve real problems through engineering, AI, and automation.
As codesPACT continues expanding its Infrastructure & AI line, you can expect more systems like this—practical, scalable, and designed for real-world applications.
Stay tuned for more project releases, sprint updates, and technical insights from our team.
