Experience
A timeline of research collaborations and industry roles focused on applied AI, LLM systems, and enterprise-grade automation.
AI Intern
Finetuned browser-use agents to autonomously query analytics platforms such as Looker Studio and Hootsuite. Built dynamic graph-based agents for factory anomaly detection and evaluated computer-vision pipelines for retail intelligence, including speaker diarisation for sales-call analytics.
Research Intern
Designed MetaSearch, a search-augmented agent for consensus resolution in peer review. Built argumentation pipelines to analyse 900+ ICLR reviews with high agreement (Cohen's Kappa = 0.834) and devised Cognitive Load heuristics for information-dense reviewing.
AI Intern
Authored enterprise whitepapers on LLM-driven requirement generation, benchmarking DeepSeek R1, GPT-4o, Gemini 2.0, and LLaMA 3.2 across BLEU, Levenshtein, and Jaccard metrics. Automated Promptflow RAG workflows, optimising chunking to reduce retrieval perplexity by 15%.
Research Assistant
Investigated C-to-Rust transpilation with existing tools and LLMs, categorising translation bugs and strengthening quality evaluation frameworks for future research.
Python (LLM) Development Intern
Built AI-powered marketing content pipelines, evaluated multiple image-generation models, and tailored GenAI prompts to create platform-specific, brand-aligned campaigns.