Daksheth Gokul

I’m a final-year Computer Science undergraduate specializing in AI & ML, focused on building practical systems that reliably understand, retrieve, and transform unstructured text. I enjoy turning complex pipelines into workflows that are easier to run, validate, and improve at scale.

At Greynium Technologies (OneIndia), I engineered multilingual news intelligence pipelines supporting 10+ Indian languages. I automated scraping, enrichment, and analysis across 5,000+ daily news articles, and implemented multilingual tag-generation and Named Entity Recognition workflows using XLM-RoBERTa to improve metadata extraction and classification accuracy across 10+ datasets and 50K+ text samples.

I also orchestrated 15+ agentic AI workflows in n8n to automate data ingestion, trigger ML pipelines, validate outputs, and reduce manual tagging effort across high-volume news streams. On top of that, I developed LLM-powered content sensitivity detection, virality prediction, and Indic English translation using Prompt Engineering, RAG architecture, and Pinecone retrieval across 10+ multilingual news sources.

Beyond core AI/ML, I’ve contributed to user-facing experiences as a Front-End Intern, collaborating with a team of 5+ engineers to build a product search engine with dynamic filtering and real-time retrieval across 1,000+ listings. I bring that same attention to quality into my projects too—like my quantum-enhanced object detection framework that achieved strong CAM classification results on COD10K-v3, and my Random Forest network anomaly detection system delivering 99% intrusion detection accuracy.