I’m building a high-quality dataset of English articles and blog posts and need every sentence fully annotated. For each piece of text you will: • mark overall sentiment (positive, negative, neutral) • tag all named entities using standard NER labels (PER, ORG, LOC, etc.) • assign part-of-speech tags to every token I will provide the raw articles plus a brief annotation guideline; you return the completed annotations in JSON or CoNLL-style files along with a short read-me describing your workflow and any custom scripts you used. Accuracy and consistency matter more than speed, so please highlight your experience with sentiment analysis, NER, POS tagging or any NLP toolkits (SpaCy, NLTK, Stanford CoreNLP, flair). If you have questions about edge cases, let’s clarify them early—I want the final dataset ready for model training without extra cleanup.