Tokenless

Tokenless

How KV Cache Speeds Up LLMs for Faster AI Models on GPUs

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
Frontier results, on device - RL Nabors, Arize

Frontier results, on device - RL Nabors, Arize

RL Nabors discusses the significant costs associated with using frontier AI models, covering security, latency, and financial implications. She introduces a framework for right-sizing AI solutions by leveraging smaller, task-specific models and Small Language Models (SLMs). The framework details how to prove task feasibility, establish success criteria with golden datasets, conduct capability evaluations (using tools like Phoenix), and select the most appropriate "Small And Good Enough" (SAGE) model. Nabors further demonstrates how prompt engineering, particularly few-shot prompting, and post-processing can close performance gaps with larger models, while advocating for continuous regression evaluations to maintain performance integrity. The overarching message is to "prototype big, deploy small" to optimize AI deployments.

Research to Reality: Bringing Frontier ML Research to Production - Vaidas Razgaitis, Higharc

Research to Reality: Bringing Frontier ML Research to Production - Vaidas Razgaitis, Higharc

Vaidas Razgaitis, Senior Research Engineer at Higharc, shares three tactical tips to accelerate the transition of novel AI/ML research into production-ready features. He emphasizes addressing the critical handoff challenge between ML researchers and software engineers through structured documentation (Research Prototype Taxonomy Document), a well-organized monorepo utilizing decoupled microservices, and a systematic approach to code decomposition and PR review. These strategies aim to improve legibility, maintainability, and delivery speed for ML-driven products.

Uncertainty-Guided Data Augmentation for Engineers | Deep Dive - Yongmin Kwon

Uncertainty-Guided Data Augmentation for Engineers | Deep Dive - Yongmin Kwon

This session details a data-efficient method for training engineering surrogate models by using uncertainty quantification (UQ) to guide geometric data augmentation. Instead of random deformations, the approach lets the deep ensemble model identify its own knowledge gaps (epistemic uncertainty), then uses Free-Form Deformation (FFD) to generate new shapes specifically in those uncertain regions. This ensures every expensive simulation run yields maximally informative data, significantly improving model accuracy for a fixed computational budget across domains like structural mechanics and aerodynamics.

Artificial Intelligence

View All
How KV Cache Speeds Up LLMs for Faster AI Models on GPUs

How KV Cache Speeds Up LLMs for Faster AI Models on GPUs

LLMs often slow down under heavy traffic due to inefficient GPU memory management during inference. This overview explains how KV cache and Paged Attention, implemented in VLLM, optimize memory usage across prefill and decode phases, significantly boosting LLM throughput, reducing latency, and improving GPU utilization through advanced context handling and specific tuning techniques like prefix caching and speculative decoding.

The AI Agents Helping Home Services Book More Jobs

The AI Agents Helping Home Services Book More Jobs

Avoca (YC W23) has achieved eight-figure revenue and a $1 billion valuation by building an AI workforce for home services, turning missed calls into revenue. Founders Apurva Shrivastava and Tyson Chen explain how AI expands software's market share beyond 1% by automating labor and operational costs, leading to a 15x larger opportunity. They emphasize that their AI agents augment human workers, reducing attrition in challenging CSR roles and creating new positions for training AI, driven by a deep customer obsession learned at YC.

What Is AI Code Review? Fixing Slow PRs & Broken Workflows with AI

What Is AI Code Review? Fixing Slow PRs & Broken Workflows with AI

Anna Gutowska explains how AI code review enhances software development by addressing the slowness and inconsistency of traditional methods. The video delves into the benefits of AI in accelerating reviews, improving code quality, fostering developer learning, and reducing technical debt. It covers the underlying technologies like static/dynamic analysis and LLMs, discusses critical considerations such as over-reliance and context, and provides best practices for integrating AI while emphasizing the indispensable role of human oversight.

Technology

View All
Platforms: Build Abstractions, not Illusions • Gregor Hohpe • GOTO 2025

Platforms: Build Abstractions, not Illusions • Gregor Hohpe • GOTO 2025

Gregor Hohpe explains the critical role of platforms in managing the growing cognitive load on developers due to complex distributed systems. He contrasts platforms, driven by "economies of speed" and fostering innovation through diversity, with traditional IT services and oversimplified abstractions that create dangerous illusions. Hohpe emphasizes building platforms that provide intuitive, domain-specific abstractions to solve real business problems, rather than just repackaging existing cloud services.

Full Stack Greenfield Projects : Are they still relevant?

Full Stack Greenfield Projects : Are they still relevant?

Bharat Goenka, co-founder of Tally, discusses the company's unconventional approach to software development through "Full Stack Greenfield" projects. He explains why building every component from scratch, despite being a high-risk strategy, has been crucial for Tally's success in serving the SMB market, fostering extreme customer loyalty, and aspiring to connect 200 million businesses. The talk delves into the historical context, the philosophy of questioning and choosing constraints, and the distinction between product and custom engineering.

3‑2‑1 Backup Rule Explained: Protect Your Data from Disaster

3‑2‑1 Backup Rule Explained: Protect Your Data from Disaster

Jeff Crume outlines essential data resiliency strategies, starting with the 3-2-1 backup rule—three copies, two media types, one offsite—and expanding to include immutable or air-gapped backups, rigorous testing, and encryption. He emphasizes these principles for robust disaster recovery, ransomware protection, and minimizing costly downtime, highlighting the trade-offs in achieving high availability.

Recent Post

The Four Types of Memory Every AI Agent Needs — with Richmond Alake

The Four Types of Memory Every AI Agent Needs — with Richmond Alake

Richmond Alake from Oracle explains the critical role of Agent Memory in building adaptive AI agents, detailing the four types of memory, the limitations of Retrieval-Augmented Generation (RAG), and the architecture of the modern agent stack.

Running LLMs on your iPhone: 40 tok/s Gemma 4 with MLX — Adrien Grondin, Locally AI

Running LLMs on your iPhone: 40 tok/s Gemma 4 with MLX — Adrien Grondin, Locally AI

Adria Grondin, developer of the Locally AI app, provides a technical walkthrough on running large language models like Google's Gemma on an iPhone using Apple's MLX framework. The talk covers the necessary tools, performance expectations, the importance of quantization, and the growing MLX ecosystem.

Full Workshop: Build Your Own Deep Research Agents - Louis-François Bouchard, Paul Iusztin, Samridhi

Full Workshop: Build Your Own Deep Research Agents - Louis-François Bouchard, Paul Iusztin, Samridhi

This hands-on workshop details the construction of a sophisticated, dual-part AI system for producing high-quality technical content. It begins with an MCP-powered deep research agent that autonomously plans, searches the web, and analyzes sources like YouTube to synthesize a grounded research artifact. The second part is a constrained, deterministic writing workflow that transforms this research into polished, non-sloppy content using an innovative "Evaluator-Optimizer" pattern for iterative refinement. The session emphasizes crucial AI engineering principles, such as choosing between agentic and workflow-based architectures, and concludes with a deep dive into implementing practical observability and evaluation pipelines to ensure the system is both measurable and improvable.

Can we AI our way to a more sustainable world?

Can we AI our way to a more sustainable world?

Microsoft experts Doug Burger, Amy Luers, and Ishai Menache discuss the dual role of AI in sustainability. They analyze the environmental footprint of datacenters and explore how AI-driven optimization and materials discovery can be pivotal in decarbonizing global systems like energy, industry, and food production.

Gemma, DeepMind's Family of Open Models — Omar Sanseviero, Google DeepMind

Gemma, DeepMind's Family of Open Models — Omar Sanseviero, Google DeepMind

A deep dive into Google DeepMind's Gemma 4, the latest family of open models. This summary covers the new model architectures like per-layer embeddings, on-device agentic capabilities, multimodal features, and the growing ecosystem of fine-tuned applications from medicine to sovereign AI.

The New Application Layer - Malte Ubl, CTO Vercel

The New Application Layer - Malte Ubl, CTO Vercel

Malte Ubl, CTO of Vercel, posits that AI engineering is the successor to web development, arguing that AI agents will expand, not shrink, the software market. He explores practical agent archetypes being built today and discusses the profound shift required in infrastructure and security as agents become both the builders and primary users of software, concluding that the true innovation and value will lie in the application layer built by AI engineers, independent of the commoditizing foundational models.

← PreviousNext →

Stay In The Loop! Subscribe to Our Newsletter.

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