I will be your expert ai data annotator and quality rater
About this Gig
Need to ensure your AI or LLM (Large Language Model) responses are accurate, relevant, and safe?
I am a Data Analyst and AI Quality Specialist with hands-on experience (at Mindrift) evaluating and improving AI model performance. I provide structured, actionable feedback based on complex guidelines.
My Services:
- AI Model Evaluation: Rigorous testing of your LLM's responses against your guidelines.
- Fact-Checking & Verification: In-depth verification of AI-generated statements.
- Data Annotation & Labeling: Meticulous annotation of text and other data to train your model.
- Quality Assurance (QA) Monitoring: Identifying bugs, inconsistencies, and safety violations.
Why Hire Me?
- Relevant AI Experience: Current, hands-on experience as an AI Agent Assistant.
- Certified: I hold the IBM Machine Learning Certificate.
- Analytical Background: Data Analyst skilled in SQL, Python, & BI tools (Power BI, Tableau).
- Bilingual: Native Spanish and C1 (Fluent) English, perfect for global projects.
I am ready to help you improve the quality and accuracy of your AI project.
Contact me or place your order to get started!
Programming language:
Python
•
R
•
SQL
Frameworks:
Scikit-learn
•
PyTorch
•
Panda
Tools:
Jupyter Notebook
•
Excel
•
Colab
•
RStudio
My Portfolio
FAQ
What kind of AI models can you evaluate?
I specialize in evaluating Large Language Models (LLMs) for tasks like chatbot responses, content generation, and summarization. My expertise is in fact-checking, safety, relevance, and adherence to complex guidelines.
Can you work with guidelines in Spanish?
Yes! I am a native Spanish speaker and C1 (Fluent) in English. I am perfectly comfortable evaluating, fact-checking, or annotating projects in either language.
What is the difference between your Basic and Standard packages?
The Basic package is great for a quick quality check (evaluation of 50 outputs). The Standard package is better for a deeper analysis, as it includes data preprocessing and performance monitoring for 120 outputs.

