RAGEkansh VermaDamian Kelly25 September 2026
/research
Insights and blogs from the AI Group building Fin
Ashwini Shukla7 May 2026Structured, Agentic RAG for Ecommerce
James O'Neill9 April 2026Low-Rank Key Value Attention: Reducing KV Cache Memory and Maintaining Head Diversity
Text ClassificationMariia Matskevichus30 March 2026Unsupervised Learning Meets Generative AI: Topic Modelling for Real-World Dialogue
PodcastFergal Reid19 September 2025Podcast EP2: Shipping reliable AI actions
- PodcastFergal Reid18 September 2025
Podcast EP1: Closing the loop
Dhruv Patel11 September 2025Building a Better Language Detection Model for Fin
RAGRamil YarullinFedor Parfenov11 September 2025Using LLMs as a Reranker for RAG: A Practical Guide
Mario Kostelac11 September 2025Building out Intercom’s AI infra
Text ClassificationVinicius Ribeiro11 September 2025“Was that helpful?” Understanding User Feedback in Customer Support AI Agents
- Benchmarking & TestingSagar JoglekarRamil Yarullin11 September 2025
David vs Goliath: are small LLMs any good?
- Text ClassificationRamil Yarullin11 September 2025
To escalate, or not to escalate, that is the question
RAGRamil Yarullin11 September 2025How We Built a World-Class Reranker for Fin
RAGDhruv Patel11 September 2025Finetuning Retrieval for Fin
- Stefan Ivanovici10 September 2025
Cost of Serving LLMs
Afonso Marques20 August 2025We Don’t Need Higher Peak Intelligence, Only More Intelligence Density
Ketan Bhatt8 August 2025Fin: Running a Reliable Service over Unreliable Parts
- Eamon Phelan18 July 2025
Think Fast: Reasoning at 3ms a Token
Fedor Parfenov7 July 2025A Causal Inference Approach to Measuring the Impact of Improved RAG Content