Tokenless

Tokenless

Stop Making Your Customers Click Approve - Emil Chetty, Wise

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
Bringing Continual Learning to Every Company: Trajectory’s Ronak Malde

Bringing Continual Learning to Every Company: Trajectory’s Ronak Malde

Ronak Malde, co-founder of Trajectory, argues that agents need to learn continuously from real-world experience instead of relying on offline batch updates. He covers how training custom models has become accessible to companies, two research directions (Group Size-1 GRPO and a meta-grader for reward propagation), the Harvey and Rogo results, the distinction between continual learning and recursive self-improvement, and a vision of product builders as gardeners who shape products that grow on their own.

Why Waiting for the Next AI Model Can Pay Off - John Stecher, Blackstone & Scott Worland, Norm AI

Why Waiting for the Next AI Model Can Pay Off - John Stecher, Blackstone & Scott Worland, Norm AI

Blackstone CTO John Stecher and Norm AI CTO Scott Worland discuss how to build AI agent systems that remain useful as models improve. They cover why data strategy and straight-through processing come first, how to weigh ROI against waiting for a better model, how evals anchor financial-services work, and why Norm rebuilt its platform after a large hand-labeled eval set exposed the cost of over-engineered code.

How We Took an AI Agent From 41% to 97% Accurate - Milind Sunil Shah, Brainbase Labs

How We Took an AI Agent From 41% to 97% Accurate - Milind Sunil Shah, Brainbase Labs

Milind Sunil Shah of Brainbase Labs presents a six-step framework for improving back-office agents, illustrated with a Domino's project that tracks competitor pizza prices across websites and delivery apps. Accuracy rose from 41% to 97%, cost per source fell from $2 to $1, and run time dropped from 129 to 71 minutes, all through instructions, skills, and playbooks rather than a new model. The talk covers kickoff, ontology, benchmarks and evals, a narrow first agent, iteration, and handing off the full loop.

Technology

View All
What iPhone Packaging Reveals About the Robotics Opportunity

What iPhone Packaging Reveals About the Robotics Opportunity

Tutor Intelligence CEO Josh Gruenstein explains why robotics is a systems problem spanning AI, hardware, manufacturing, and customers, and how deploying robots into real factories and warehouses, with human tutors as backup, generates the data that makes them more capable. The discussion covers the company's own robot designs, its focus on small and mid-sized businesses, the economics of flexible robotic labor, and the implications for American manufacturing and supply chains.

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.

Recent Post

Distant conversational speech recognition: Challenges and Opportunities

Distant conversational speech recognition: Challenges and Opportunities

Dr. Samuele Cornell from Carnegie Mellon University discusses the persistent challenges in distant automatic speech recognition (DASR) for spontaneous, multi-party conversations. He explains why state-of-the-art systems falter in real-world scenarios and presents recent advancements through three key efforts: (1) insights from the CHiME-7/8 DASR challenges, which benchmark robust meeting transcription; (2) progress towards unified end-to-end models that jointly handle diarization and recognition; and (3) novel techniques for generating realistic, large-scale training data using a combination of large language models and multi-speaker text-to-speech systems.

IronDict: Transparent Dictionaries from Polynomial Commitments

IronDict: Transparent Dictionaries from Polynomial Commitments

Hossein Hafezi from NYU presents IronDict, a novel transparent dictionary construction using polynomial commitment schemes. IronDict addresses the major limitations of existing Merkle tree-based systems, such as high auditing costs and imperfect privacy. By modeling the dictionary with polynomials and leveraging the algebraic properties of the KZH commitment scheme, IronDict achieves perfect privacy and dramatically reduces auditing overhead, making it feasible for end-users to verify the system's integrity on consumer devices.

Lattice-Based Accumulator and Application to Anonymous Credential Revocation

Lattice-Based Accumulator and Application to Anonymous Credential Revocation

Victor Youdom Kemmoe from Brown University presents a novel, communication-efficient cryptographic accumulator based on the Module-SIS assumption. This construction is designed for applications like anonymous credential revocation, where elements can be added without needing to update existing membership witnesses, a significant improvement over previous post-quantum schemes.

Efficient Secure Aggregation for Federated Learning

Efficient Secure Aggregation for Federated Learning

Varun Madathil from Yale University presents Tacita, a novel, single-server protocol for secure aggregation in Federated Learning (FL). Tacita is designed to address the unique constraints of the FL environment, such as client dropouts and the absence of client-to-client communication. The protocol achieves one-shot execution with constant-size communication and robustness against dropouts by introducing two new cryptographic primitives: succinct multi-key linearly homomorphic threshold signatures (MKLHTS) and a homomorphic variant of Silent Threshold Encryption.

Overcoming Agentic Memory Management Challenges

Overcoming Agentic Memory Management Challenges

Biswaroop Bhattacharjee from Prem AI discusses Cortex, a novel AI memory system inspired by human cognition. The conversation explores moving beyond traditional flat memory structures to hierarchical, context-aware systems that enable more sophisticated and less noisy retrieval for AI agents.

How To Be Contrarian — And Right

How To Be Contrarian — And Right

Garry, Harj, Jared, and Diana discuss why founders should pursue contrarian ideas in a crowded AI market. They analyze how companies like Uber, Coinbase, and Flock Safety found massive success by tackling non-obvious, legally ambiguous, or seemingly impossible problems that others ignored.

← PreviousNext →

Stay In The Loop! Subscribe to Our Newsletter.

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