Devesh S.

Key Skills

Software

Other

Top Subject Matter

Supervised ML training data annotation (NLP: intent, NER, sentiment)

Operational data preparation and anomaly flagging for ML-ready datasets

Top Data Types

Text

Image

Top Task Types

Data Collection

Freelancer Overview

Data Annotation & Labelling Specialist — Innodata Inc. (2022–2024). Brings 2+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal, Proprietary Tooling, and Other. Education includes Post Graduate Diploma in Management, G.L. Bajaj Institute of Management & Research (2026) and Bachelor of Science, University of Lucknow (2021). AI-training focus includes data types such as Text and labeling workflows including Entity (NER) and Data Collection.

Labeling Experience

Data Operations & MIS Intern — Einfratech Systems (Jun–Jul 2025)

OtherTextData Collection

Conducted data quality and risk analysis using systematic anomaly detection to reduce entry errors and improve dataset cleanliness. Coordinated real-time data updates with cross-functional teams to support reliable monitoring of operational and pipeline data. Produced performance reports summarizing outlier trends to guide process improvement recommendations. • Applied anomaly detection approaches to flag and reduce problematic entries. • Supported dataset curation for downstream ML training readiness. • Collaborated with teams to maintain accurate and timely data updates. • Generated outlier-focused reports for iterative data/process refinement.

2025 - 2025

Data Annotation & Labelling Specialist — Innodata Inc. (2022–2024)

Text

Annotated and labeled large-scale multimodal datasets, including text, to support supervised ML model training for enterprise clients. Performed NLP and semantic tagging for intent classification, named entity recognition (NER), and sentiment labeling tasks. Maintained annotation accuracy above 97% while following detailed guidelines and quality benchmarks for reliable training data. • Labeled intents, entities (NER), and sentiment for downstream model training. • Refined labeling schemas and improved inter-annotator agreement through QA collaboration. • Handled edge cases by coordinating with cross-functional QA teams. • Ensured consistency and adherence to annotation standards across projects.

2022 - 2024

Education

Work History