Insights

Notes from the build.

Field notes on custom AI, legal intelligence, and how we ship tools our clients own — including sourced, interactive breakdowns of the research we read.

Latest
Research Breakdown

Harness²: Recursive Agent Harnessing for an Open World

Xu, Chen, CuiZhu et al. · Northwestern University, Google Cloud AI Research

Published October 20, 2026 · 3 min read + 5 min videoRead breakdown
Research Breakdown

HarnessDev: Can LLMs Create and Evolve Their Own Agent Harness?

Wu, Zhang, Shi et al. · ByteDance Seed, SUTD, Georgia Tech, M-A-P, TokenWave.AI

Published October 15, 2026 · 3 min read + 4 min videoRead breakdown
Research Breakdown

False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents

Chen, Li, Lu et al. · Rutgers, UC San Diego, Michigan, McGill, KFUPM

Published October 15, 2026 · 3 min read + 4 min videoRead breakdown
Research Breakdown

Raven: The Harness of Harnesses for Composable Agentic Intelligence

EverMind AI

Published October 15, 2026 · 3 min read + 3 min videoRead breakdown
Research Breakdown

JEV-as-a-Judge: Accept When Confident, Escalate When Unsure

Yubo Li et al. · Carnegie Mellon University

Published October 15, 2026 · 3 min read + 3.5 min videoRead breakdown
Research Breakdown

Reimagining research papers as interactive and reliable AI agents

Jiacheng Miao et al. · Stanford University

Published October 15, 2026 · 3 min read + 4 min videoRead breakdown
Research Breakdown

Crossing the Reward Bridge: Expanding RL with Verifiable Rewards Across Diverse Domains

Yi Su et al. · Tencent AI Lab · Soochow University

Published September 23, 2026 · 9 min readRead breakdown
Research Breakdown

The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity

Parshin Shojaee et al. · Apple

Published September 16, 2026 · 9 min readRead breakdown
Research Breakdown

Retentive Network: A Successor to Transformer for Large Language Models

Yutao Sun et al. · Microsoft Research · Tsinghua University

Published September 9, 2026 · 10 min readRead breakdown
Research Breakdown

Distilling the Knowledge in a Neural Network

Geoffrey Hinton et al. · Google Inc.

Published September 2, 2026 · 9 min readRead breakdown
Research Breakdown

LoRA: Low-Rank Adaptation of Large Language Models

Edward Hu et al. · Microsoft Corporation

Published August 26, 2026 · 10 min readRead breakdown
Research Breakdown

Learning to summarize from human feedback

Nisan Stiennon et al. · OpenAI

Published August 13, 2026 · 10 min readRead breakdown
Research Breakdown

Chain-of-Thought Prompting Elicits Reasoning in Large Language Models

Jason Wei et al. · Google Research, Brain Team

Published August 12, 2026 · 10 min readRead breakdown
Research Breakdown

Improving Language Understanding by Generative Pre-Training

Alec Radford et al. · OpenAI

Published August 5, 2026 · 10 min readRead breakdown
Research Breakdown

BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

Jacob Devlin et al. · Google AI Language

Published July 29, 2026 · 10 min readRead breakdown
Research Breakdown

Attention Is All You Need

Ashish Vaswani et al. · Google Brain · Google Research · U. Toronto

Published July 22, 2026 · 9 min readRead breakdown
Interactive

How SIFTY Got Smart

SIFTAI · A SIFTY origin story

Published July 15, 2026 · 10 stages · ~8 minRead breakdown

More breakdowns on the way.

We publish sourced breakdowns and field notes as we build. Have a paper you want us to dig into, or a problem of your own? Let's talk.

Email UsBook a Call