Manasjyoti Gogoi

I’m an AI Data Operations and LLM Evaluation specialist with hands-on experience evaluating AI-generated video, voice-assistant, and multimodal agent outputs using structured rubrics, evidence-based rationales, and project-specific quality criteria.

At Innodata, I perform RLHF-style pairwise ranking, prompt and response evaluation, ASR transcription, vision-language annotation, multimodal AI-agent evaluation, and hierarchical robotic-action captioning across text, image, audio, and video workflows. I’ve maintained 95%+ accuracy on Multi Mango evaluation tasks and earned a 4.67/5 QA audit score for robotic-action captioning. I also conduct Tier-3 quality reviews, investigate edge cases, and document detailed evidence to support consistent evaluation decisions.

Alongside AI data work, I have experience with Python, SQL, Power BI, Excel, data analysis, and dashboard development. I’m currently expanding my skills in machine learning, generative AI, and AI engineering, with a long-term focus on building reliable AI systems and evaluation pipelines.

I also have a background in C# and Unity, having independently developed a 3D project whose trailer and promotional content reached 100K+ YouTube views.