M

Muskan Pareek

About

Detail

Bengaluru, Karnataka, India

Contact Muskan regarding: 
Flexible work
Starting at USD15/hour

Timeline


work
Job
school
Education
folder
Project

Résumé


Jobs verified_user 0% verified
  • Defined.ai
    AI Data Trainer
    Defined.ai
    Feb 2024 - Current (2 years 7 months)
    • Trained AI models by creating and refining prompts for large language models (LLMs). • Reviewed and evaluated AI-generated outputs for accuracy, relevance, and safety. • Annotated and structured datasets for NLP and conversational AI training pipelines. • Collaborated with AI engineers to improve model performance through feedback loops. • Conducted quality assurance checks on training datasets and AI responses.
  • iMerit
    AI Data Annotator
    iMerit
    Jul 2023 - Jan 2024 (7 months)
    • Labeled and categorized large-scale datasets used for machine learning and NLP models. • Evaluated prompt-response pairs to enhance LLM accuracy and contextual understanding. • Assisted in building high-quality datasets for AI model training and evaluation. • Maintained strict data quality standards through multi-layer validation processes.
Education verified_user 0% verified
  • G
    Bachelor of Technology (B.Tech) in Computer Science and Engineering
    Govt. Women Engineering College, Ajmer Rajasthan
    Aug 2021 - Jun 2025 (3 years 11 months)
Projects (professional or personal) verified_user 0% verified
  • C
    Customer Sentiment Analysis using NLP
    Jan 2024 - May 2024 (5 months)
    • Built a Python-based model to classify customer reviews using NLP techniques. • Used Pandas, Scikit-learn, and NLP preprocessing.
  • S
    Sales Data Dashboard
    Aug 2024 - Dec 2024 (5 months)
    • Created an interactive Power BI dashboard analyzing sales trends and forecasting. • Integrated and cleaned data from multiple sources using Excel and SQL, enabling real-time tracking of key performance indicators (KPIs) such as revenue growth, regional sales performance, and product demand.
  • A
    AI Prompt Optimization System
    May 2025 - Sep 2025 (5 months)
    • Designed and tested prompts to improve LLM response accuracy. • Reduced irrelevant responses by 30% through prompt refinement.