Tokenless

Tokenless

Why We Rebuilt Microsoft Office From Scratch for AI Agents - Yijia Liang, Compound AI

Interactive discovery

Explore the topic map

Follow the connections between themes, people, and ideas across the Tokenless archive in an interactive topic modeling map.

Open the map Browse posts

Machine Learning

View All
Judge a Transaction by the Company It Keeps — Somya Rai & MD Sharique, EXL

Judge a Transaction by the Company It Keeps — Somya Rai & MD Sharique, EXL

EXL developed an advanced fraud detection system that augments XGBoost with Graph Neural Network (GNN) embeddings. Traditional tabular models struggle to capture fraud signals residing in transaction neighborhoods. The system's innovative design includes meticulous sampling strategies, disjoint monthly graph construction for leakage control, and a GraphSAGE architecture that aggregates mean, max, and standard deviation of neighbor features to detect subtle fraud patterns. Emphasizing explainability for regulatory compliance and analyst trust, it also features a highly optimized serving architecture achieving 3ms P99 latency, pivoting to edge prediction to reduce cache footprint from 60GB to 2-3GB. This led to a 27-33% relative lift in true positive rates and a projected $65 million incremental fraud capture.

From 15% to 90% GPU Utilization: Fix the Data Pipeline, Not the Model

From 15% to 90% GPU Utilization: Fix the Data Pipeline, Not the Model

Tarun Sunkaraneni from Amazon AGI shows how a multimodal SFT pipeline that spent about 85% of its time waiting on data was brought to roughly 90% GPU utilization. He walks through concurrency with asyncio and Ray actors, prefetching, zero-copy transfer via Ray's object store, and a network-card bottleneck that appears only at scale, and argues that the hardest part of training throughput is often keeping GPUs fed rather than optimizing the model.

Small Models, Big Results: Training a Finance Agent for Under $500 — Charles Dickens, Snorkel AI

Small Models, Big Results: Training a Finance Agent for Under $500 — Charles Dickens, Snorkel AI

Snorkel AI and UC Berkeley's Sky Computing Lab show that a 4B Qwen3 model, trained with reinforcement learning for under $500 using the rLLM framework, beats its 235B sibling on expert-validated financial question answering built from SEC 10-K filings. The talk covers the FinQA dataset construction and verification, common failure modes of frontier models, transfer to multi-table and general tool-use tasks, and ablations showing that simple data and binary rewards outperform more elaborate setups, since the real bottleneck is disciplined tool use rather than reasoning depth.

Artificial Intelligence

View All
Why We Rebuilt Microsoft Office From Scratch for AI Agents - Yijia Liang, Compound AI

Why We Rebuilt Microsoft Office From Scratch for AI Agents - Yijia Liang, Compound AI

Yijia Liang, founder of Compound and former finance analyst, explains why personal agents succeeded and what finance needs beyond them: multiplayer agents with their own inboxes, a headless Office engine rebuilt with agents, and agent swarms that navigate massive data rooms and models while keeping every output grounded in citations.

Most of Your Docs Readers Are Now AI Agents - Hahnbee Lee, Mintlify & Lincoln Murr, Coinbase

Most of Your Docs Readers Are Now AI Agents - Hahnbee Lee, Mintlify & Lincoln Murr, Coinbase

Hahnbee Lee of Mintlify and Lincoln Murr of Coinbase discuss what changes when AI agents become the main users of software products. They cover why documentation now needs to be optimized for context windows and accuracy rather than design, how Coinbase is exposing its business through APIs, and how the x402 payment standard and crypto microtransactions let agents pay for services and gain greater autonomy.

If Your AI Scores 100%, Your Eval Is Wrong - Ramakrishnan Lokanathan, eslams.com

If Your AI Scores 100%, Your Eval Is Wrong - Ramakrishnan Lokanathan, eslams.com

Ramakrishnan Lokanathan argues that AI evals are broken because they are overcomplicated, and lays out a practical framework for building them for high-stakes applications: define the most ambitious version of success, test every step of a workflow, combine automated checks with human review and adversarial tests, guard against regressions, and turn real user failures into tests. He also introduces eSlams, an open-source package where models compete in games like chess, Go, and poker as a window into how they reason.

Technology

View All
Why Founders Are Betting Big on Factories, Mines, and Energy

Why Founders Are Betting Big on Factories, Mines, and Energy

Erin Price-Wright and Connor Love unpack the AD50 and the rebuilding of America's industrial base, covering how defense demand is financing manufacturing capacity for commercial markets, how AI is reshaping factory economics, how a new generation of founders applies lessons from SpaceX, Tesla, Anduril, and Palantir to mining, energy, and manufacturing, and why private capital is now willing to fund physical infrastructure before government demand is guaranteed.

Mark Cuban’s Plan to Make Healthcare Affordable

Mark Cuban’s Plan to Make Healthcare Affordable

Mark Cuban and Julie Yoo discuss how transparency, direct contracting, and AI could reshape healthcare economics, from Cost Plus Drugs and modular pod manufacturing to personalized medicine, the future of physicians, and advice for founders who want to use AI to learn rather than to stop thinking.

Exploits Now Beat Patches: Open Source Security in the AI Era - Patrick Smyth, Chainguard

Exploits Now Beat Patches: Open Source Security in the AI Era - Patrick Smyth, Chainguard

Patrick Smyth of Chainguard examines how frontier AI has compressed the time between vulnerability disclosure and exploitation, leaving 30-day remediation cycles obsolete. The talk covers AI-assisted attacks, waves of supply chain malware, four attack surfaces spanning malware, known CVEs, zero days, and side doors like CI/CD, developer laptops, and agent skills, and practical defenses including cooldown periods, faster CVE remediation, and signing artifacts with SLSA and Sigstore.

Recent Post

Connecting the Dots with Context Graphs — Stephen Chin, Neo4j

Connecting the Dots with Context Graphs — Stephen Chin, Neo4j

Stephen Chin of Neo4j argues that traditional RAG is insufficient because AI agents lose the reasoning behind past decisions. He introduces Context Graphs as a solution to capture the 'why' behind decisions, creating a queryable system of precedent that provides grounded, explainable, and auditable results.

Agents Don't Do Standups: Building the Post-Engineer Engineering Org — Mike Spitz, PFF

Agents Don't Do Standups: Building the Post-Engineer Engineering Org — Mike Spitz, PFF

A case study from PFF reveals how a two-engineer team, by leveraging AI agents, achieved a 25x increase in deployment frequency and 10x the output of a ten-engineer team. CTO Mike Spitz explains their core principle: shifting focus from making engineers faster to making AI agents faster. This talk deconstructs the resulting transformation, which eliminated traditional agile ceremonies like stand-ups and sprint planning in favor of an automated, spec-to-PR workflow, redefining the roles of engineers and processes in a modern software organization.

Combine Skills and MCP to Close the Context Gap — Pedro Rodrigues, Supabase

Combine Skills and MCP to Close the Context Gap — Pedro Rodrigues, Supabase

Pedro Rodrigues from Supabase shares key lessons from building an agent skill to work with Postgres and Supabase. He explains why critical security rules must go in the main skill file, the importance of pointing to living documentation, and how providing opinionated workflow guidance closes the reliability gap for agents in production systems.

How Building with AI Can Double the Throughput of Your Engineering Team — Brian Scanlan, Intercom

How Building with AI Can Double the Throughput of Your Engineering Team — Brian Scanlan, Intercom

Intercom doubled its engineering throughput in under a year by treating its AI coding agent not as a simple tool, but as a new senior engineer. This involved a full onboarding process onto their 15-year-old Rails monolith, creating a library of durable skills for recurring tasks, and providing audited access to all internal systems.

Ye Cannae Change the Laws of Physics • Kevlin Henney • GOTO 2025

Ye Cannae Change the Laws of Physics • Kevlin Henney • GOTO 2025

Kevlin Henney explores the limits of software abstraction, arguing that while software is an 'executable fiction,' it is ultimately constrained by the fundamental laws of physics. This talk deconstructs common metaphors like 'velocity' and 'roadmap' and delves into the real-world implications of physical constants like the speed of light and impossibility theorems like CAP.

Tokenmaxxing vs AI Hardware Bottlenecks — with Jon Krohn (@JonKrohnLearns)

Tokenmaxxing vs AI Hardware Bottlenecks — with Jon Krohn (@JonKrohnLearns)

While the 'tokenmaxxing' trend grows, the AI industry faces severe physical infrastructure bottlenecks. This summary explores the four key constraints choking AI compute: GPU packaging (CoWoS), high-bandwidth memory (HBM), the surprising surge in CPU demand from agentic AI, and critical electricity shortages, revealing how these challenges are shaping the future of AI development.

← PreviousNext →

Stay In The Loop! Subscribe to Our Newsletter.

Get updates straight to your inbox. No spam, just useful content.