Open to opportunities

Rakshita Goudadi

@rakshitagoudadi

Message

AI Data Quality professional driving annotation accuracy and scalable QA for NLP and generative AI.

Ireland

Message

What I'm looking for

I seek a role where I can lead annotation quality, build scalable QA frameworks, mentor teams, and drive ethical AI practices within fast-paced, data-driven environments.

I am an AI Data Quality Analyst with 3+ years leading annotation and QA operations for NLP, computer vision, and LLM training support. I design SOPs, run calibration and coaching, and build dashboards that translate data into actionable quality improvements.

At Telus International I led multimedia annotation QA, authored SOPs, produced 50+ root cause analyses, and improved guideline clarity while delivering training that boosted QA scores. At iMerit I annotated text, audio, and image datasets with 98%+ accuracy and authored guidelines that increased team QA pass rates by 80%.

My technical toolkit includes Labelbox, Playment, GT-Studio, CODA, Tableau, Python, SQL, Java, and advanced Excel, which I apply to measure F1, inter-annotator agreement, error trends, and calibration outcomes. I mentor teams, optimize workflows, and partner with cross-functional groups to improve model performance.

I am passionate about ethical AI, bias mitigation, and operational excellence in high-volume environments, and I bring a data-driven, people-centered approach to delivering reliable datasets at scale.

Experience

Work history, roles, and key accomplishments

Quality Analyst

Telus International

Dec 2023 - Dec 2024 (1 year)

Led QA for multimedia annotation workflows, authored SOPs and performed 50+ root cause analyses that reduced error rates and increased annotation accuracy for production datasets.

Root Cause Analysis Dashboarding Labelbox GT Studio Tableau)

Data Annotator

iMerit Technology & Services

Jul 2021 - Jul 2023 (2 years)

Annotated text, audio, and image datasets with 98%+ accuracy, authored guidelines that raised team QA pass rates by 80% and contributed to a 15% improvement in ML model performance.