Tokenless

Tokenless

When Running Your Own AI Models Beats Paying for APIs - Steven Latré, imec

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
THE $800B COMPUTE TRAP

THE $800B COMPUTE TRAP

Ben Lorica and Evangelos Simoudis work through Anthropic's S-1 filing, weighing its striking growth and improving unit economics against compute commitments, falling prices from open-weight models, and dependence on coding. They then examine SpaceX's emerging neocloud business and close with a debate over proposed frontier AI regulation, from self-improvement limits to independent testing and incident disclosure.

When Running Your Own AI Models Beats Paying for APIs - Steven Latré, imec

When Running Your Own AI Models Beats Paying for APIs - Steven Latré, imec

Steven Latré, who leads AI at the semiconductor research center imec, presents a benchmarking study of what it really costs to move long-running coding agents from commercial APIs to self-hosted GPUs. Across thousands of SWE-bench Pro runs, the study finds that hardware costs rise in steps rather than smoothly, that concurrency limits matter more than raw accuracy, that break-even GPU utilization ranges from about 89% on small systems to about 15% on 8x B200 machines, and that harness choice can halve GPU time for the same results.

Voice AI Where One Wrong Question Breaks the Rules - Shiv Indap, Upstart & Brooke Hopkins, Coval

Voice AI Where One Wrong Question Breaks the Rules - Shiv Indap, Upstart & Brooke Hopkins, Coval

Shiv Indap of Upstart and Brooke Hopkins of Coval walk through automating loan verification calls in a heavily regulated lending environment. They cover the compliance traps that shape the agent's behavior, moving human review beyond a Google Sheet, and raising LLM judge agreement with human reviewers from 33% to 92%. The talk then lays out a continuous loop of live monitoring, human review queues, simulation-based testing of fixes, and self-healing agents, with humans signing off on every change.

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

Semantic Blindness: 500,000 Sensors Confused an LLM - Raahul Singh & Vanč Levstik, Phaidra

Semantic Blindness: 500,000 Sensors Confused an LLM - Raahul Singh & Vanč Levstik, Phaidra

Modern LLMs struggle with combinatorial engineering problems in industrial settings due to "Semantic Blindness" – an inability to understand physical system topology, scale efficiently, or handle repetitive naming conventions. This talk introduces a hybrid AI architecture that combines LLMs for high-level planning with deterministic systems for execution, leveraging hierarchical structures and pattern-based search. This approach achieves 100% accuracy and flat operational costs at massive scales, demonstrating an "inversion" of the Software 1.0/3.0 paradigm where AI-native systems mature by integrating deterministic code for reliable, structured tasks.

The Agentic Web and the Bazaar Era of AI - Ramesh Raskar, MIT Media Lab

The Agentic Web and the Bazaar Era of AI - Ramesh Raskar, MIT Media Lab

Project Nanda introduces the Agentic Web, an open infrastructure designed to break AI agents out of "walled gardens" and enable truly interconnected, autonomous interactions. It focuses on three core layers: Discovery, using the Nanda Index for agents to find and trust each other; Commerce, facilitating "knowledge pricing" markets for agents to pay for intelligence and services; and the Bazaar, fostering machine co-learning through the Nanda Town simulator, a sandbox for testing large-scale agent coordination and protocols. This initiative aims to establish an open, permissionless ecosystem for trillions of AI agents.

Why the tech workforce is quietly splitting in two | Annual AI sentiment survey (Noam Segal)

Why the tech workforce is quietly splitting in two | Annual AI sentiment survey (Noam Segal)

Noam Segal and Lenny Rachitsky's second annual Tech Worker Sentiment Survey reveals a striking bifurcation in the tech workforce: 50% feel energized and amplified by AI, while the other half feels redefined, destabilized, or diminished. Burnout is surging (up 11 points in a year) and optimism is declining, even as 97% believe AI makes them 'better' (often meaning faster, not higher quality, leading to 'cognitive rot'). The #1 fear isn't job loss to AI, but being squeezed to do more for the same pay. Nobody would recommend their job to someone entering the industry today, with designers and researchers being the most negative. Managers are identified as the single biggest lever for employee well-being, yet most are rated ineffective. The industry is described as 'chaotic,' with advice for employees to go deep on AI, manage the 'squeeze,' and invest in manager relationships, while leaders are urged to invest in managers, manage expectations, and support early talent and struggling roles.

Stop Evaluating Models Like It's the 50s - Alejandro Vidal, Mindmakers

Stop Evaluating Models Like It's the 50s - Alejandro Vidal, Mindmakers

This talk introduces the application of psychometrics, particularly Item Response Theory (IRT), to improve LLM evaluation. It highlights how IRT goes beyond simple accuracy to measure model ability, item difficulty, and discrimination, enabling benchmark auditing, adaptive testing, data leakage detection, and the identification of model relationships and distillation, ultimately providing deeper insights into what LLMs truly learn.

Claws Out: Securing and Building with OpenClaw - Nick Taylor, Pomerium

Claws Out: Securing and Building with OpenClaw - Nick Taylor, Pomerium

This talk details how the speaker contributed a "trusted-proxy auth mode" to the OpenClaw project to enhance security and user experience by eliminating the need for tokens and device pairing. It explains the technical implementation of this mode using Identity-Aware Proxies like Pomerium. The speaker also demonstrates a unique workflow, building and live-editing a Multi-Modal Chatbot Protocol (MCP) application directly within ChatGPT, controlled by their OpenClaw instance via Discord, showcasing the power of secure, natural language-driven web development.

Stop AI Agent Hallucinations: 5 Techniques + Production Patterns - Elizabeth Fuentes, AWS

Stop AI Agent Hallucinations: 5 Techniques + Production Patterns - Elizabeth Fuentes, AWS

Explore five research-backed techniques to overcome common AI agent architectural limitations like hallucination, inefficient tool use, and rule violations. Learn how to implement semantic tool selection, Graph-RAG, multi-agent validation, neurosymbolic guardrails, and agent steering through code changes, not prompt engineering, and deploy them using AWS Bedrock Agent Core for robust and cost-effective AI agents.

← PreviousNext →

Stay In The Loop! Subscribe to Our Newsletter.

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