Maria C.

Key Skills

Software

No software listed

Top Subject Matter

AI agents for customer support and service automation

Top Data Types

Text

Top Task Types

Function Calling

Transcription

Text Summarization

Text Generation

Question Answering

Object Detection

Freelancer Overview

AI agent training, evaluation, and continuous improvement (prompting/QA). Brings 7+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Make, n8n, and ManyChat. Education includes Degree, N/A (2026). AI-training focus includes data types such as Text and labeling workflows including Evaluation and Rating.

Labeling Experience

AI Agents and Automation Specialist - N/A

I designed, trained, evaluated, and continuously improved AI agents and automation workflows for real business environments. I focused on refining agent performance through output analysis, prompt iteration, and knowledge base adjustments to improve accuracy and reliability. I ensured interactions in customer support and service contexts were high-quality, contextually appropriate, and guideline-compliant for better user experience. • Built and deployed AI-powered agents and automation workflows using Make, n8n, and ManyChat • Analyzed agent outputs to identify inconsistencies and failure patterns, then iterated on instructions and prompts • Developed and tested edge cases to improve robustness and reduce ambiguous responses • Performed response evaluation and quality assurance using performance data and real-world feedback

2022 - Present

AI agent training, evaluation, and continuous improvement (prompting/QA)

Text

Trained and evaluated AI agents and automation workflows by analyzing outputs for inconsistencies, improving prompts, and refining knowledge bases in order to increase accuracy and reliability. Performed iterative quality assurance on customer support/service interactions by reviewing agent responses and identifying failure patterns to improve instruction sets. Continuously adjusted agent behavior using real-world interaction feedback to ensure contextually appropriate, guideline-following outputs. • Reviewed agent responses and assessed quality against expected behavior • Identified inconsistencies and recurring failure patterns from interaction data • Refined prompts and instruction sets based on performance evidence • Tested edge cases and iterated to improve accuracy and user experience

2022 - Present

AI Agent Developer - N/A

I built AI agent behavior by translating user needs into reliable instruction-following and information-retrieval workflows. I evaluated agent responses for quality, reliability, and usability, then refined system prompts and knowledge resources to address weaknesses. I applied human judgment to review outputs, assess quality, and deliver iterative improvements based on real interaction patterns. • Improved agent instruction sets to strengthen guideline adherence and reduce misinterpretation • Refined knowledge bases and prompt strategies to handle ambiguity and improve contextual responses • Reviewed responses and provided feedback to support continuous behavior tuning • Tested performance against edge cases to increase trustworthiness and user satisfaction

2020 - 2021

Education

Work History