Tokenless

Tokenless

Why Humans Can't Keep Up With AI-Written Code - Tariq Shaukat, 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
The Cheaper AI Model That Costs You More - Rayan Krishnan, Vals AI

The Cheaper AI Model That Costs You More - Rayan Krishnan, Vals AI

Rayan Krishnan of Vals AI explains why single benchmark scores in model cards hide a jagged frontier, using finance tasks like tax and Excel LBO modeling to show that the best model changes by task and that cheaper models can cost more. He covers benchmark design lessons (accuracy metrics, validity, reliability, reward hacking, multi-turn work), a case study in which a team cut token spend about 90% using internal evals built from its own GitHub history, and why a firm's own work history is its most valuable eval source.

Catch the Bugs Your Code Review Misses - Killian Carlsen-Phelan & Taylor Luttrell-Williams, Sonar

Catch the Bugs Your Code Review Misses - Killian Carlsen-Phelan & Taylor Luttrell-Williams, Sonar

A hands-on walkthrough of Gitar, a code change verification agent, applied to three pull requests in a sample Python ticketing app. It covers catching a bug outside the diff, enforcing a business rule through project configuration, and diagnosing a failing CI test before auto-merging, along with functional validation against Jira, learned conventions, and cross-repo and cross-PR analysis.

Why Humans Can't Keep Up With AI-Written Code - Tariq Shaukat, Sonar

Why Humans Can't Keep Up With AI-Written Code - Tariq Shaukat, Sonar

Sonar CEO Tariq Shaukat argues that agents have made code abundant but confidence scarce, and draws on Toyota's quality-first factory principles to show how to build software factories enterprises can trust. He covers five hard problems that break agentic development and presents Sonar's code confidence formula: intent, guide rails, verification, and evidence.

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

Why smarter AI models could drive up compute prices 10x

Why smarter AI models could drive up compute prices 10x

The author analyzes the looming imbalance between rapid AI lab revenue growth and slower compute capacity expansion. He explores how this dynamic will likely drive up compute costs and favor highly efficient, frontier models, creating significant barriers to entry and reshaping the AI landscape in the coming years.

Agentic Engineering vs Software Engineering: Beyond Vibe Coding

Agentic Engineering vs Software Engineering: Beyond Vibe Coding

Anna Gutowska explains the paradigm shift in software engineering towards "agentic engineering," where AI agents execute goals defined by developers. She differentiates this from traditional, AI-assisted, and vibe coding, highlighting the increased importance of human oversight, orchestration, and verification in a world of probabilistic AI systems, and how this redefines the developer's role.

MCP Apps: Extending the Frontier — Ido Salomon & Liad Yosef

MCP Apps: Extending the Frontier — Ido Salomon & Liad Yosef

MCP Apps transforms chat and coding assistant interactions from text-heavy responses into rich, interactive user interfaces, preserving brand identity and enhancing user experience. It allows services to send "UI atoms" directly into assistants like Claude and ChatGPT, enabling a "write once, run anywhere" model across a massive user base. This standard defines how hosts render web components, manage user interactions, and shifts control of the user journey to the agent, ushering in a new era of the agentic web.

MCP Tasks (async): Why Aren't Any Agents Supporting Them? — Cornelia Davis, Temporal

MCP Tasks (async): Why Aren't Any Agents Supporting Them? — Cornelia Davis, Temporal

Cornelia Davis explains why the MCP tasks specification, designed for durable, long-running, and interactive operations, hasn't seen widespread adoption. She details V1's architectural flaws, particularly concerning scalability and stateful human-in-the-loop interactions, and introduces V2's shift to a stateless core, client-initiated updates, and a clearer path for client-side durability, while outlining future work on scaling and adoption.

When Will The Benchmaxxing Plague End? — Nick Heiner, Surge AI

When Will The Benchmaxxing Plague End? — Nick Heiner, Surge AI

Nick Heiner explores the phenomenon of "benchmaxing" in AI, where models are optimized for benchmark scores rather than real-world utility. He exposes common antipatterns in benchmark creation, such as contamination, reward hacking, and misaligned verifiers, and critiques labs' tactics like gaming leaderboards. Heiner advocates for a higher standard, emphasizing the need for human expertise, high-fidelity data, and rigorous alignment in evaluation to ensure benchmarks genuinely reflect AI's value.

This CPO regrets that product management exists | Tom Verrilli (CPO of Whatnot)

This CPO regrets that product management exists | Tom Verrilli (CPO of Whatnot)

Tom Verrilli, CPO at Whatnot, discusses his controversial stance that "we regret product management exists," advocating for a leaner, more hands-on PM approach. He delves into how Whatnot structures its product team, its rigorous hiring process focused on deep systems thinking and IC work, and the transformative impact of AI on product and data science functions. Verrilli also shares mental models for balancing strategy and iteration, navigating the CPO-founder relationship, and key lessons from his time at Twitter, emphasizing the importance of core PM skills in an AI-driven world.

← PreviousNext →

Stay In The Loop! Subscribe to Our Newsletter.

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