Weekday AI is an emerging artificial‑intelligence startup focused on building multilingual vision‑language models that can understand complex documents across diverse scripts. The company’s mission is to democratise AI by creating high‑quality training data for low‑resource languages, enabling products that work equally well for Hindi, Telugu, Japanese, Korean and other scripts. With a small but highly motivated team of data scientists, linguists and engineers, Weekday AI partners with global tech firms to deliver datasets that power OCR, document‑understanding and conversational AI solutions. The role of PDF Annotation & Transcription Expert – Telugu is a remote, contract‑based position that forms the backbone of Weekday AI’s data‑generation pipeline. As a native Telugu speaker, you will work directly on real‑world PDF pages – ranging from newspaper spreads and school exam sheets to flyers and manuals – and transform them into richly annotated, machine‑readable training examples. Every annotation, component type, reading order and transcription you produce will be reviewed by a second expert, ensuring the highest level of quality before the data is fed into model training. Key responsibilities include: 1. Verify each assigned PDF page for language, legibility and absence of personal data. 2. Identify and bound every meaningful region (titles, headings, paragraphs, lists, tables, figures, diagrams, captions, formulas, question/answer fields) using the provided annotation tool. 3. Assign correct component types and a logical reading‑order index that mirrors how a human would read the page. 4. Link child components to their parent figures or tables via parent identifiers. 5. Transcribe every text region verbatim in Telugu script, preserving diacritics, conjuncts and handwritten nuances; flag any illegible sections. 6. Capture page‑level metadata such as language, document type, source, dimensions and special flags for tables, formulas and handwriting. 7. Review the work of fellow annotators, providing feedback and ensuring consistency across the dataset. 8. Continuously apply the taxonomy and guidelines, maintaining systematic consistency across hundreds of pages. 9. Report any unsuitable pages early to avoid wasted effort. 10. Contribute suggestions for improving annotation guidelines based on on‑ground experience. The tech stack is lightweight – a web‑based annotation interface, PDF viewer, and basic spreadsheet tools for metadata capture. Familiarity with Unicode Telugu input, OCR post‑editing and data‑labeling platforms is advantageous. Growth opportunities include moving into senior annotation lead, quality‑assurance specialist, or even data‑annotation project manager as the dataset scales. Joining Weekday AI offers the chance to work on cutting‑edge AI research, collaborate with an international team, and make a tangible impact on language technology for Telugu speakers. Why join? You will gain hands‑on experience with AI data pipelines, improve your linguistic precision, and enjoy the flexibility of remote work while contributing to a socially valuable mission of language inclusion.