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

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 transformative potential of Item Response Theory (IRT), a psychometric model, for evaluating AI models. It addresses the limitations of current evaluation methods by demonstrating how IRT can precisely measure model intelligence, audit and optimize benchmarks, detect data leakage, implement adaptive testing, identify item bias, and even fingerprint models based on their error patterns, opening new avenues for understanding and improving LLM capabilities.

From fork() to Fleet: Designing an Agent Sandbox Cloud — Abhishek Bhardwaj, OpenAI

From fork() to Fleet: Designing an Agent Sandbox Cloud — Abhishek Bhardwaj, OpenAI

This talk by Abhishek Bhardwaj from OpenAI details the architectural considerations for building secure and scalable AI agent sandboxes in the cloud. It explores runtime isolation technologies (from basic process execution to containers, GVisor, and microVMs), emphasizing the superior security of hardware-virtualized microVMs. The speaker then highlights the critical need for persistent storage, outlining explicit (copy-on-write snapshots) and always-on (tiered block storage) solutions as the next major unlock for agent capabilities. Finally, it touches on orchestration challenges for fleet-level management, including low-latency sandbox creation and snapshot-driven scheduling.

The Prime Intellect Stack — Will Brown, Prime Intellect

The Prime Intellect Stack — Will Brown, Prime Intellect

Deep dive into Prime Intellect's open-source ecosystem for post-training LLMs, covering the modular Verifiers V1 environment design, the asynchronous and scalable Primer RL training framework, and the Lab platform for hosted training and fine-tuning. Learn about advanced reward systems, the interception server pattern, and tokenization control with the Renderers library, all designed to enable frontier agentic model development.

What is an AI Code Generator? LLM Coding, Productivity, & Risk

What is an AI Code Generator? LLM Coding, Productivity, & Risk

AI code generators leverage LLMs to translate natural language into code, significantly boosting developer productivity and job satisfaction by automating boilerplate and accelerating learning. However, they introduce risks like subtle security vulnerabilities and require rigorous human review. Evaluating tools hinges on trust, demanding features like data provenance, governance, secure deployment, and curated training data for enterprise adoption.

RLM: Recursive Language Models for Large Codebases - Shashi, Superagentic AI

RLM: Recursive Language Models for Large Codebases - Shashi, Superagentic AI

This talk introduces Recursive Language Models (RLM) as a solution for coding agents struggling with large codebases. RLM externalizes context management into a programmable REPL, allowing the model to write code, inspect the repository, and recursively delegate sub-questions via `llm_query` to curate relevant context. The speaker demonstrates `RLM Code`, an independent implementation, showcasing its operational loop, observability features, and practical applications for AI engineers dealing with complex software projects.

The AI bugpocalypse is here. Now what? - Jack Cable, Corridor

The AI bugpocalypse is here. Now what? - Jack Cable, Corridor

Jack Cable discusses the "AI bug apocalypse" driven by advanced AI models finding and exploiting vulnerabilities and AI coding tools increasing attack surfaces. He champions a "secure by design" approach, advocating for systemic changes like using memory-safe languages to prevent common vulnerability classes rather than just patching. He also addresses AI's role in introducing new vulnerabilities, the shift towards autonomous AI in development, and policy recommendations for securing the future of AI-powered coding.

← PreviousNext →

Stay In The Loop! Subscribe to Our Newsletter.

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