Deshant P.

Key Skills

Software

Don't disclose

Top Subject Matter

Bilingual translation quality evaluation (English-Hindi)

Bilingual content annotation for sentiment/tone/factual accuracy

Text data preprocessing and quality flagging for language tasks

Top Data Types

Text

Top Task Types

Classification

Evaluation/Rating

Transcription

Prompt + Response Writing (SFT)

Object Detection

Freelancer Overview

Translation Quality Evaluation Exercise (English ↔ Hindi). Core strengths include Don't disclose, Internal, and Proprietary Tooling. Education includes Bachelor of Technology, Maharishi Markandeshwar (Deemed to be University) (2029) and Higher Secondary Certificate, Shemford Futuristic School (2024). AI-training focus includes data types such as Text, Computer Code, and Programming and labeling workflows including Evaluation, Rating, and Translation.

Labeling Experience

Basic Text-Processing Automation Script (Python)

Built an automated text-processing workflow to clean and standardize raw text data before downstream labeling or evaluation. Implemented duplicate removal, encoding fixes, and formatting normalization to reduce manual preprocessing effort. Added simple string-matching to flag phrases that may be mistranslated or out of context. • Cleaned and standardized raw text (removed duplicates, fixed encoding, normalized formatting). • Reduced manual cleanup time by ~60% on a 5,000-line sample. • Used string-matching logic to flag potentially mistranslated/out-of-context phrases. • Supported preparation of data for subsequent annotation or quality evaluation tasks.

2026 - 2026

Bilingual Content Annotation Project (Python, Spreadsheets)

Text

Annotated a bilingual text dataset for sentiment, tone, and factual accuracy across two languages to support consistent language evaluation. Applied a Python script to organize and validate annotation consistency throughout the dataset. Authored annotation guidelines to ensure shared labeling standards across reviewers. • Annotated 500 sentences across Language 1 and Language 2. • Labeled for sentiment, tone, and factual accuracy. • Used Python for organization and annotation consistency checks. • Documented labeling guidelines for consistency across reviewers.

2026 - 2026

Translation Quality Evaluation Exercise (English ↔ Hindi)

Don't discloseText

Conducted translation quality evaluation for English↔Hindi outputs by reviewing machine-translated sentences for linguistic and contextual correctness. Identified errors related to grammar, tone, and cultural context, then rewrote flawed translations to improve fidelity and natural phrasing. Produced a report summarizing common failure patterns observed during the review. • Reviewed a set of 100 machine-translated sentences. • Flagged issues such as grammar errors, tone mismatches, and cultural/contextual mistranslations. • Rewrote 30+ flawed translations to better preserve meaning. • Documented recurring error types including idiom mistranslation, gender agreement issues, and literal translation problems.

2026 - 2026

Education

Work History