Wineberry Logo
Client LoginWork With Us

Papers We Love.

The research that shapes how we think about building and deploying AI systems in production.

Attention Is All You Need

Vaswani et al. · NeurIPS 2017

The foundational architecture behind every modern LLM. Understanding attention mechanisms is non-negotiable for applied AI work.

TransformersArchitecture

Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks

Lewis et al. · NeurIPS 2020

The paper that formalized RAG — the pattern we use most frequently in production deployments for grounding LLM outputs.

RAGRetrieval

Constitutional AI: Harmlessness from AI Feedback

Bai et al. · Anthropic 2022

Critical reading for anyone building AI systems that need to operate within safety constraints at scale.

SafetyAlignment

ReAct: Synergizing Reasoning and Acting in Language Models

Yao et al. · ICLR 2023

The reasoning + acting paradigm that underpins how we design agentic workflows — think before you tool-call.

AgentsReasoning

Scaling Laws for Neural Language Models

Kaplan et al. · OpenAI 2020

Understanding scaling laws helps us advise clients on model selection and the economics of fine-tuning vs. prompting.

ScalingTraining

LoRA: Low-Rank Adaptation of Large Language Models

Hu et al. · ICLR 2022

The go-to technique for efficient model adaptation. We use LoRA and its variants in nearly every private deployment.

Fine-tuningEfficiency

This is a living list. We update it as new research shapes our practice.