Tokenless

Tokenless

Your AI Agent Starts Every Session Like a New Hire - Killian Carlsen-Phelan, Sonar

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
How to Decide What Your AI Agent Should Remember - Wojciech Pluta, Oracle

How to Decide What Your AI Agent Should Remember - Wojciech Pluta, Oracle

Wojciech Pluta, head of technical marketing for Oracle AI Database, argues that RAG gives agents a read path while memory gives them a write path, and that the write path is where business value lives. He explains memory promotion, a pipeline that admits, rejects, or escalates candidate memories, and demonstrates it with a mortgage agent that remembers preferences across sessions while refusing unsafe rules. The talk also covers short-term, long-term, and shared memory, why databases are a strong storage layer, and the context card that keeps prompts lean.

Your AI Agent Starts Every Session Like a New Hire - Killian Carlsen-Phelan, Sonar

Your AI Agent Starts Every Session Like a New Hire - Killian Carlsen-Phelan, Sonar

Killian Carlsen-Phelan of Sonar introduces Sonar Vortex, a plugin that gives coding agents a map of a codebase and its guidelines before they write, then deterministically verifies every edit. In a live demo, it catches a temp-file permissions bug that the agent's own tests missed and rejects the agent's first fix.

Please Stop Telling Your AI to “Search Harder”

Please Stop Telling Your AI to “Search Harder”

A discussion of how agent-driven traffic overwhelms APIs, gateways, and cloud infrastructure built for predictable load. It compares microservices with agent workloads, revisits SaaS multi-tenancy, and examines Agent Substrate's actor and worker model for suspend-and-resume agent sessions, with a critical view of whether most users will ever need this layer directly.

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

GPT-Red: Can AI read teams stop prompt injections?

GPT-Red: Can AI read teams stop prompt injections?

This podcast explores AI's impact on cybersecurity, discussing OpenAI's GPT-Red for automated red-teaming against prompt injections, the open-source ScamBuster that uses AI to bait scammers for threat intelligence, and Bruce Schneier's insights on the widening gap between skill and ability in the AI era and its ethical implications for the field.

Better Agent Auth — Bereket Habtemeskel & Paola Estefania, Better Auth

Better Agent Auth — Bereket Habtemeskel & Paola Estefania, Better Auth

The Agent Auth protocol introduces a novel approach to securing AI agents by treating them as distinct principals with their own identities, moving away from sharing user credentials. It defines a system for agent discovery, fine-grained capability-based authorization, and comprehensive traceability, enabling users to grant specific authority to agents and revoke it granularly. This paradigm shift offers enhanced security, accountability, and control over autonomous operations.

Causal Models Need Causal Data - Xaira’s X-Cell model (Bo Wang & Ci Chu)

Causal Models Need Causal Data - Xaira’s X-Cell model (Bo Wang & Ci Chu)

Bo Wang and Ci Chu of Xaira Therapeutics introduce X-Cell, a 4.9-billion-parameter diffusion language model designed to predict cellular responses to genetic perturbations. Trained on PISCES, the largest genome-wide CRISPRi Perturb-seq dataset (25.6 million single cells across 16 contexts), X-Cell overcomes limitations of previous observational models. The discussion highlights the critical role of high-quality, causal data, X-Cell's architecture advantages, and its demonstrated ability to generalize to unseen cell types and primary cells, marking a significant step towards truly predictive virtual cell models in drug discovery.

2026 State of AI Engineering — Barr Yaron, Amplify Partners

2026 State of AI Engineering — Barr Yaron, Amplify Partners

Barr Yaron's 2026 AI engineering survey reveals key trends: audio and image generation are rapidly gaining traction, while cost is now a primary engineering constraint. Agents are evolving to take actions within systems, but control mechanisms remain primitive. Evaluation (eval) is still the top infrastructure challenge. AI positively impacts job satisfaction and experimentation but also raises concerns about technical skill erosion and non-developers shipping code, fundamentally changing engineering culture. Predictions include a likely AGI declaration within five years and a shift away from Transformers as state-of-the-art.

Why Physical AI Is the Next Frontier | The a16z Show

Why Physical AI Is the Next Frontier | The a16z Show

Applied Intuition discusses the emergence of physical AI, its mission to put intelligence on a billion machines, and its latest platform, Dana, designed to democratize autonomous system development. The conversation covers the vast scope of physical AI beyond automotive, the unique challenges of real-world deployment (safety, data, hardware), the current state and future of self-driving cars and trucks, and the transformative potential of humanoids and world models. They also touch upon the geopolitical landscape of AI and the global ambitions of Applied Intuition.

WebAssembly on Kubernetes • Nicolas Frankel • YOW! 2025

WebAssembly on Kubernetes • Nicolas Frankel • YOW! 2025

Nicolas Fränkel explores the evolution of WebAssembly (Wasm) beyond its web origins, showcasing its potential to revolutionize application deployment on Kubernetes. The talk demonstrates how Wasm enables incredibly small container sizes (down to 2MB for an HTTP server) by integrating specific Wasm runtimes with Kubernetes' extensible architecture. However, Fränkel also provides a candid assessment of the ecosystem's rapid, often unstable, development, recommending Wasm on Kubernetes for agile startups seeking competitive advantage but cautioning traditional enterprises due to the inherent risks and maintenance challenges.

← PreviousNext →

Stay In The Loop! Subscribe to Our Newsletter.

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