Bridging the gap between human linguistics and artificial intelligence. I bring 6+ years of experience in high-volume localization, LLM output evaluation, AI training and precise prompt engineering.
Transcreation, MTPE, AIPE, Mastery in LQA, terminology management, style guide development, and cultural adaptation. Native Hindi, fluent English & Urdu.
Rigorous model output evaluation, training dataset preparation, RLHF annotation, and advanced prompt engineering.
Leveraging code to streamline language processing, manage massive data extraction, and automate complex QA workflows.
When a client provided a new 700-term glossary a day before a 7,000-word banking localization deadline, manual updates were unfeasible. I wrote a Python script to compare the glossaries, isolate the 73 changed terms, and automatically update 111 instances across the main file in 30 minutes.
Read PostAdapted English subtitles of a Chinese micro-drama into natural Hindi for dubbing. By carefully matching the Hindi dialogue to the original actors' emotions, pauses, and lip movements, the voice actors were able to record the script in record time with a highly natural delivery.
Read PostAnalyzed Hindi outputs across major AI models (ChatGPT, Claude, Gemini) and identified a systemic failure in generating correct nuqtas and chandrabindus. Highlighted how models adopt inconsistent web content, underscoring the need for rigorous LQA in AI training datasets.
Read PostExploring how Artificial Intelligence acts as a great equalizer, empowering small businesses and freelancers to compete with larger entities by automating complex workflows and scaling operations efficiently.
A critical look at how AI generation impacts artistic integrity, specifically examining the trend of 'Ghiblification' and whether algorithmic imitation can capture the true essence of human creativity and emotion.