AI Quality Assurance specialist & Audio Annotator — bridging precision data evaluation with the nuance of human perception to build better AI systems.
I specialize in making AI systems more reliable and human-aligned. As an AI QA specialist, I evaluate model outputs for quality, coherence, and safety — providing the kind of precise feedback that helps machine learning teams improve their models meaningfully.
On the audio side, I annotate and transcribe speech data for ASR and TTS training pipelines, with a sharp ear for phonetics, tone, and contextual accuracy — including Indonesian and multilingual datasets.
I bring both analytical rigor and a human perspective to every task, understanding that good AI starts with good data.
Performed side-by-side evaluation of two AI chatbot models sharing identical input capabilities but differing in their internal data processing architectures. Assessed each model's ability to correctly interpret and act on user inputs — including real-world mobile device interactions such as app-launching commands. Identified and documented failure patterns where one model consistently fell short in applying user intent. Completed 300+ evaluation queues with structured comparative feedback.
Annotated 500+ audio files by identifying and labeling both verbal speech — transcribed phonetically as spoken — and non-verbal human sounds such as laughter, crying, and other paralinguistic cues, marked using standardized bracket notation.
Recorded 300+ audio-visual files of short sentences in both Indonesian and English, each delivered with specific emotional expressions. The dataset was designed to train AI models on local pronunciation patterns and emotion-aware speech recognition.
Open to freelance projects in AI data annotation, QA evaluation, and audio transcription. I bring precision, reliability, and a genuine care for data quality to every collaboration.
Send Me an Email →