AI | Techlist.io

googleAug 5, 2025Original article

Insulin resistance prediction from wearables and routine blood biomarkers (opens in new tab)

Researchers at Google have developed a novel machine learning approach to predict insulin resistance (IR) by integrating wearable device data with routine blood biomarkers. This method aims to provide a scalable, less invasive alternative to traditional "gold standard" tests like the euglycemic insulin clamp or specialized HOMA-IR assessments. The study demonstrates that combining digital biomarkers with common laboratory results can effectively identify individuals at risk for type 2 diabetes, particularly within high-risk populations. ## Barriers to Early Diabetes Screening * Insulin resistance is a primary precursor to approximately 70% of type 2 diabetes cases, yet it often remains undetected until the disease has progressed. * Current diagnostic standards are frequently omitted from routine check-ups due to high costs, invasiveness, and the requirement for specific insulin blood tests that are not standard practice. * Early detection is vital because insulin resistance is often reversible through lifestyle modifications, making accessible screening tools a high priority for preventative medicine. ## The WEAR-ME Multimodal Dataset * The research utilized the "WEAR-ME" study, which collected data from 1,165 remote participants across the U.S. via the Google Health Studies app. * Digital biomarkers were gathered from Fitbit and Google Pixel Watch devices, tracking metrics such as resting heart rate, step counts, and sleep patterns. * Clinical data was provided through a partnership with Quest Diagnostics, focusing on routine blood biomarkers like fasting glucose and lipid panels, supplemented by participant surveys on diet, fitness, and demographics. ## Predictive Modeling and Performance * Deep neural network models were trained to estimate HOMA-IR scores by analyzing different combinations of the collected data streams. * While models using only wearables and demographics achieved an area under the receiver operating characteristic curve (auROC) of 0.70, adding fasting glucose data boosted the auROC to 0.78. * The most comprehensive models, which combined wearables, demographics, and full routine blood panels, achieved the highest accuracy across the study population. * Performance was notably strong in high-risk sub-groups, specifically individuals with obesity or sedentary lifestyles. ## AI-Driven Interpretation and Literacy * To assist with data translation, the researchers developed a prototype "Insulin Resistance Literacy and Understanding Agent" built on the Gemini family of large language models. * The agent is designed to help users interpret their IR risk predictions and provide personalized, research-backed educational content. * This AI integration aims to facilitate better communication between the data results and actionable health strategies, though it is currently intended for informational and research purposes. By utilizing ubiquitous wearable technology and existing clinical infrastructure, this approach offers a path toward proactive metabolic health monitoring. Integrating these models into consumer or clinical platforms could lower the barrier to early diabetes intervention and enable more personalized preventative care.

AI Large Language Models Machine Learning Gemini+4

googleAug 5, 2025Original article

Highly accurate genome polishing with DeepPolisher: Enhancing the foundation of genomic research (opens in new tab)

DeepPolisher is a deep learning-based genome assembly tool designed to correct base-level errors with high precision, significantly enhancing the accuracy of genomic research. By leveraging a Transformer architecture to analyze sequencing data, the tool reduces total assembly errors by 50% and insertion or deletion (indel) errors by 70%. This advancement is critical for creating near-perfect reference genomes, such as the Human Pangenome Reference, which are essential for identifying disease-causing variants and understanding human evolution. ## Limitations of Current Sequencing Technologies * Genome assembly relies on reading nucleotides (A, T, G, and C), but the microscopic scale of these base pairs makes accurate, large-scale sequencing difficult. * Short-read sequencing methods provide high signal strength but are limited to a few hundred nucleotides because identical DNA clusters eventually desynchronize, blending signals together. * Long-read technologies can sequence tens of thousands of nucleotides but initially suffered from high error rates (~10%); while tools like DeepConsensus have reduced this to 0.1%, further refinement is necessary for high-fidelity reference genomes. * Even a 0.1% error rate results in millions of inaccuracies across the 3-billion-nucleotide human genome, which can cause researchers to miss critical genetic markers or misidentify proteins. ## DeepPolisher Architecture and Training * DeepPolisher is an open-source pipeline adapted from the DeepConsensus model, utilizing a Transformer-based neural network. * The model was trained using a human cell line from the Personal Genomes Project that is estimated to be 99.99999% accurate, providing a "ground truth" for identifying and correcting errors. * The system takes sequenced bases, their associated quality scores, and the orientation of the DNA strands to learn complex error patterns that traditional methods might miss. * By combining sequence reads from multiple DNA molecules of the same individual, the tool iteratively "polishes" the assembly to reach the accuracy required for reference-grade data. ## Impact on Genomic Accuracy and Gene Discovery * The tool’s ability to reduce indel errors by 70% is particularly significant, as these specific errors often interfere with the identification of protein-coding genes. * DeepPolisher has already been integrated into major research efforts, including the enhancement of the Human Pangenome Reference, providing a more robust foundation for clinical diagnostics. * Improved assembly accuracy allows for better mapping of regions where the genome is highly repetitive, which were previously difficult to sequence and assemble confidently. For researchers and bioinformaticians, DeepPolisher represents a vital step in moving from "draft" genomes to high-fidelity references. Adopting this tool in assembly pipelines can drastically improve the reliability of variant calling and gene annotation, especially in complex clinical and evolutionary studies.

AI Deep Learning Transformer Genomics+4

googleJul 31, 2025Original article

MLE-STAR: A state-of-the-art machine learning engineering agent (opens in new tab)

MLE-STAR is a state-of-the-art machine learning engineering agent designed to automate complex ML tasks by treating them as iterative code optimization challenges. Unlike previous agents that rely solely on an LLM’s internal knowledge, MLE-STAR integrates external web searches and targeted ablation studies to pinpoint and refine specific pipeline components. This approach allows the agent to achieve high-performance results, evidenced by its ability to win medals in 63% of Kaggle competitions within the MLE-Bench-Lite benchmark. ## External Knowledge and Targeted Ablation The core of MLE-STAR’s effectiveness lies in its ability to move beyond generic machine learning libraries by incorporating external research and specific performance testing. * The agent uses web search to retrieve task-specific, state-of-the-art models and approaches rather than defaulting to familiar libraries like scikit-learn. * Instead of modifying an entire script at once, the system conducts an ablation study to evaluate the impact of individual pipeline components, such as feature engineering or model selection. * By identifying which code blocks have the most significant impact on performance, the agent can focus its reasoning and optimization efforts where they are most needed. ## Iterative Refinement and Intelligent Ensembling Once the critical components are identified, MLE-STAR employs a specialized refinement process to maximize the effectiveness of the generated solution. * Targeted code blocks undergo iterative refinement based on LLM-suggested plans that incorporate feedback from prior experimental failures and successes. * The agent features a unique ensembling strategy where it proposes multiple candidate solutions and then designs its own method to merge them. * Rather than using simple validation-score voting, the agent iteratively improves the ensemble strategy itself, treating the combination of models as a distinct optimization task. ## Robustness and Safety Verification To ensure the generated code is both functional and reliable for real-world deployment, MLE-STAR incorporates three specialized diagnostic modules. * **Debugging Agent:** Automatically analyzes tracebacks and execution errors in Python scripts to provide iterative corrections. * **Data Leakage Checker:** Reviews the solution script prior to execution to ensure the model does not improperly access test dataset information during the training phase. * **Data Usage Checker:** Analyzes whether the script is utilizing all available data sources, preventing the agent from overlooking complex data formats in favor of simpler files like CSVs. By combining external grounding with a granular, component-based optimization strategy, MLE-STAR represents a significant shift in automated machine learning. For organizations looking to scale their ML workflows, such an agent suggests a future where the role of the engineer shifts from manual coding to high-level supervision of autonomous agents that can navigate the vast landscape of research and data engineering.

AI Large Language Models AI Agents Python+5

figmaJul 30, 20253 min readCurated summary

Figma’s IPO: Design Is Everyone’s Business | Figma Blog

Figma’s IPO marks a new phase for the company, but not a change in its founding mission: narrowing the gap between imagination and reality. CEO Dylan Field argues that AI will make design more accessible and important, while emphasizing that Figma will prioritize decades-long growth over short-term efficiency or share-price performance. He sees Figma as a collaborative platform where more people can shape products and ideas. ## Figma’s IPO and Long-Term Mission - Figma went public to improve corporate governance, increase brand awareness, provide liquidity, strengthen its acquisition currency, and access capital markets. - Field especially values public ownership because it allows the broader Figma community to share in the company’s success. - He cautions investors that public-market performance is unpredictable and does not promise share-price growth. - Figma will prioritize supporting designers’ evolving needs and pursuing long-term growth over maximizing quarterly efficiency. - The company expects to take significant risks, including large platform investments and mergers and acquisitions. ## AI as a Strategic Investment - Figma is investing heavily in AI and plans to increase that investment further. - This spending may reduce efficiency for several years, but Field considers AI central to the future of design workflows. - Existing capabilities, including Figma Make and other AI features, are presented as only the beginning. - AI could help designers work more effectively and bring more people into the design process. ## Design as a Competitive Advantage - Creating a minimum viable product is easier than ever, making design, craft, and distinctive perspective more important differentiators. - Design is no longer an afterthought focused only on form and function; it can determine whether a product succeeds or fails. - As design becomes more central, companies will involve more types of contributors and encourage greater experimentation and creativity. - Figma must balance accessibility with professional power while supporting collaboration and decision-making across large organizations. ## The Evolution of AI Interfaces - Field compares current AI interfaces, which rely heavily on prompts, to the MS-DOS era of computing. - He expects new, domain-specific design patterns to make AI capabilities easier and more intuitive to use. - Just as graphical user interfaces expanded access to computers, well-designed AI interfaces could make advanced capabilities available to everyday users. ## Figma’s Role in the Future - Field describes Figma as a “peaceful garden” where individuals and teams can develop ideas together. - He believes tools alone do not change the world; people use them to create meaningful change. - Figma’s long-term ambition is to help more people participate in design and turn ideas into reality. Figma’s direction is therefore one of patient, ambitious investment: accept near-term inefficiency, expand access to design, and build AI-powered tools for a much broader creative community.

Read original

AI Figma Generative AI Product Design+2

figmaJul 29, 20252 min readCurated summary

Figma Announces Pricing of Initial Public Offering | Figma Blog

Figma announced pricing for its initial public offering at **$33 per share**, valuing an offering of 36,937,080 Class A shares. Trading is expected to begin on the New York Stock Exchange under **“FIG”** on July 31, 2025, with closing scheduled for August 1. The IPO marks Figma’s transition from a design tool into a broader, AI-powered product development platform. ## IPO Details - Figma is offering **12,472,657 shares** of Class A common stock. - Existing stockholders are selling **24,464,423 shares**. - The offering’s public price is **$33.00 per share**. - Certain selling stockholders granted underwriters a 30-day option to purchase up to **5,540,561 additional shares** for over-allotments. - Figma will not receive proceeds from shares sold by existing stockholders. ## Trading and Underwriting - Shares are expected to trade on the **New York Stock Exchange** under ticker symbol **FIG**. - The offering is expected to close on **August 1, 2025**, subject to customary conditions. - Morgan Stanley, Goldman Sachs, Allen & Company, and J.P. Morgan are joint lead book-running managers. - BofA Securities, Wells Fargo Securities, and RBC Capital Markets are additional book-running managers. - William Blair and Wolfe | Nomura Alliance are serving as co-managers. - The SEC declared the related registration statement effective on July 30, 2025. ## Figma’s Platform and Evolution - Founded in 2012, Figma describes itself as a collaborative platform for digital product development. - The company has expanded beyond interface design into an AI-powered system supporting: - Ideation - Design - Building - Product shipping - Its central value proposition is helping teams collaborate more efficiently while maintaining alignment throughout the product lifecycle. Figma’s IPO makes its public-market debut with a $33-per-share offering and positions the company as a broader collaborative platform for designing and delivering digital products.

Read original

AI Figma Collaboration Product Development+1

googleJul 28, 2025Original article

Simulating large systems with Regression Language Models (opens in new tab)

Researchers from Google have introduced Regression Language Models (RLMs) as a universal solution for numeric prediction tasks by framing regression as a text-to-text problem. By converting complex, unstructured system data into strings, RLMs can predict performance metrics without the need for manual feature engineering or data normalization. This approach allows large language models to move beyond subjective human feedback and directly model raw operational data for large-scale software and industrial infrastructures. ## Conceptualizing Text-to-Text Regression * Traditional regression methods rely on tabular data—fixed-length numeric vectors—which are difficult and laborious to maintain for evolving systems like software logs or hardware patterns. * RLMs represent the input state ($x$) as a structured text string (such as JSON or YAML) and the numerical output ($y$) as a text string. * The model is trained using standard next-token prediction and cross-entropy loss, allowing it to function as a universal approximator for complex data types. * This paradigm eliminates the need for manual feature engineering, as the model learns directly from the raw textual representation of the system state. ## Architecture and Training for Large Systems * The research utilizes a compact RLM consisting of a two-layer encoder-decoder architecture with 60 million parameters. * To manage large inputs that can reach up to 1 million tokens, the system reorders features by importance at the beginning of the string so that critical data is preserved when truncated to the model's 8k token limit. * Pre-training the RLM on diverse regression tasks enables few-shot adaptation, allowing the model to adjust to new data types with minimal gradient updates. * Numerical values are processed as-is within the text, removing the requirement for traditional scaling or normalization common in standard machine learning pipelines. ## Optimizing Google's Borg Infrastructure * The method was specifically applied to Google’s Borg system to predict MIPS per GCU (Millions of Instructions Per Second per Google Compute Unit), a vital efficiency metric. * The RLM simulates the outcomes of complex bin-packing algorithms within a "digital twin" framework to optimize resource allocation across CPUs and TPUs. * By analyzing execution traces and textual metadata, the model provides high-accuracy forecasting for diverse workloads including Gmail, YouTube, and Maps. ## Density Capture and Uncertainty Modeling * Unlike traditional regressors that provide a single point estimate, RLMs can capture full probability distributions by sampling the decoded output multiple times. * This density estimation is critical for modeling aleatoric uncertainty, which represents the inherent randomness and stochastic load demands of large-scale compute environments. * The ability to visualize these distributions helps engineers identify the range of possible outcomes and the inherent variability of the system's performance over time. This research demonstrates that small, specialized language models can effectively replace traditional regression methods in highly dynamic environments. For practitioners looking to implement these capabilities, the open-source `regress-lm` library provides a framework for simulating large systems and predicting performance across varied industrial and scientific use cases.

AI Large Language Models Google Borg Regression Language Models+5

googleJul 27, 2025Original article

SensorLM: Learning the language of wearable sensors (opens in new tab)

SensorLM is a new family of foundation models designed to bridge the gap between high-dimensional wearable sensor data and natural language descriptions. By training on a massive dataset of nearly 60 million hours of de-identified health data, the models learn to interpret complex physiological signals to provide meaningful context for human activities. This research demonstrates that integrating multimodal sensor signals with language models enables sophisticated health insights, such as zero-shot activity recognition and automated health captioning, that significantly outperform general-purpose large language models. ## Dataset Scale and Automated Annotation * The models were pre-trained on an unprecedented 59.7 million hours of multimodal sensor data collected from over 103,000 individuals across 127 countries. * To overcome the high cost of manual annotation, researchers developed a hierarchical pipeline that automatically generates text descriptions by calculating statistics and identifying trends within the raw sensor streams. * Data was sourced from Fitbit and Pixel Watch devices, representing nearly 2.5 million person-days of activity and health information. ## Hybrid Training Architecture * SensorLM unifies two primary multimodal strategies: contrastive learning and generative pre-training. * Through contrastive learning, the model learns to discriminate between different states—such as a "light swim" versus a "strength workout"—by matching sensor segments to corresponding text descriptions. * The generative component allows the model to "speak" for the sensors, producing nuanced, context-aware natural language captions directly from high-dimensional biometric signals. ## Activity Recognition and Cross-Modal Capabilities * The model demonstrates state-of-the-art performance in zero-shot human activity recognition, accurately classifying 20 different activities without any specific fine-tuning. * Its few-shot learning capabilities allow the model to adapt to new tasks or individual user patterns with only a handful of examples. * SensorLM facilitates cross-modal retrieval, enabling users or experts to find specific sensor patterns using natural language queries or to generate descriptions based on specific sensor inputs. ## Generative Health Captioning * Beyond simple classification, the model can generate hierarchical captions that describe the statistical, structural, and semantic dimensions of a user’s data. * Experimental results using metrics like BERTScore show that SensorLM produces captions that are more factually correct and coherent than those created by powerful non-specialist LLMs. * This capability allows for the translation of abstract data points, such as heart rate variability or step counts, into readable summaries that explain the "why" behind physiological changes. By providing a framework where wearable data can be understood through the lens of human language, SensorLM paves the way for more intuitive and personalized health monitoring. This technology holds the potential to transform raw biometric streams into actionable insights, helping users better understand the relationship between their activities and their overall physical well-being.

AI Multimodal AI Foundation Models Wearable Technology+5

figmaJul 27, 20252 min readCurated summary

Figma Announces Increase in IPO Price Range | Figma Blog

Figma announced an increased expected price range of **$30–$32 per share** for its proposed initial public offering. The company has amended its Form S-1 registration statement and plans to list Class A common stock on the New York Stock Exchange under **“FIG.”** The announcement follows the launch of Figma’s IPO roadshow. ## Updated IPO Terms - Figma plans to offer **12,472,657 shares** of Class A common stock. - Existing stockholders plan to sell an additional **24,464,423 shares**. - Selling stockholders may grant underwriters a 30-day option to purchase up to **5,540,561 additional shares** to cover over-allotments. - Figma will not receive proceeds from shares sold by existing stockholders. ## Underwriters - Joint lead book-running managers: - Morgan Stanley - Goldman Sachs - Allen & Company - J.P. Morgan - Additional book-running managers include BofA Securities, Wells Fargo Securities, and RBC Capital Markets. - William Blair and Wolfe | Nomura Alliance will serve as co-managers. ## Regulatory Status - The offering is being made through a preliminary prospectus. - Figma’s registration statement has been filed with the SEC but is not yet effective. - Shares cannot be sold, and purchase offers cannot be accepted, until the registration statement becomes effective and applicable securities-law requirements are satisfied. ## About Figma - Founded in 2012, Figma describes itself as an AI-powered platform for collaborative product development. - Its product supports the full workflow from ideation and design through development and shipping. - The company emphasizes collaboration, efficiency, and keeping product teams aligned. Figma’s revised price range signals increased expectations for its upcoming IPO, though the final offering price and completion of the listing remain subject to market conditions and regulatory approval.

Read original

AI Figma Collaboration Product Development+1

googleJul 21, 2025Original article

LSM-2: Learning from incomplete wearable sensor data (opens in new tab)

LSM-2 introduces a paradigm shift in processing wearable sensor data by treating naturally occurring data gaps as inherent features rather than errors to be corrected. By utilizing the Adaptive and Inherited Masking (AIM) framework, the model learns directly from fragmented, real-world data streams without the need for biased imputation or data-discarding filters. This approach allows LSM-2 to achieve state-of-the-art performance in health-related classification and regression tasks, maintaining robustness even when sensors fail or data is highly interrupted. ## The Challenge of Pervasive Missingness * Real-world wearable data is almost never continuous; factors such as device charging, motion artifacts, and battery-saving modes create frequent "missingness." * Traditional self-supervised learning models require complete data, forcing researchers to use imputation—which can introduce artificial bias—or aggressive filtering that discards over 90% of potentially useful samples. * In a dataset of 1.6 million day-long windows, research found that not a single sample had 0% missingness, highlighting the impracticality of training only on complete datasets. ## Adaptive and Inherited Masking (AIM) * AIM extends the Masked Autoencoder (MAE) framework by treating "inherited" masks (naturally occurring gaps) and "artificial" masks (training objectives) as equivalent. * The framework utilizes a dual masking strategy: it employs token dropout on a fixed ratio of tokens to ensure computational efficiency during encoding. * To handle the unpredictable and variable nature of real-world gaps, AIM uses attention masking within the transformer blocks for any remaining masked tokens. * During evaluation and fine-tuning, the model relies solely on attention masking to navigate naturally occurring gaps, allowing for accurate physiological modeling without filling in missing values. ## Scale and Training Architecture * LSM-2 was trained on a massive dataset comprising 40 million hours of de-identified wearable data from more than 60,000 participants using Fitbit and Google Pixel devices. * The model learns to understand underlying physiological structures by reconstructing masked segments across multimodal inputs, including heart signals, sleep patterns, and activity levels. * Because it is trained on fragmented data, the resulting foundation model is significantly more resilient to sensor dropouts in downstream tasks like hypertension prediction or stress monitoring. LSM-2 demonstrates that foundation models for health should be built to embrace the messiness of real-world environments. By integrating missingness directly into the self-supervised learning objective, developers can bypass the computational and statistical overhead of imputation while building more reliable diagnostic and monitoring tools.

AI Transformer Foundation Models Wearable Technology+3

figmaJul 20, 20252 min readCurated summary

Launching the roadshow for Figma’s proposed IPO | Figma Blog

Figma announced the launch of its roadshow for a proposed initial public offering of Class A common stock. The offering includes new shares from Figma and shares sold by existing stockholders, with an expected price of $25–$28 per share. Figma plans to list on the New York Stock Exchange under “FIG,” pending regulatory effectiveness. ## Proposed Share Offering - Figma will offer 12,472,657 shares of Class A common stock. - Existing stockholders will offer 24,646,423 additional shares. - Selling stockholders may grant underwriters a 30-day option to purchase up to 5,540,561 more shares for over-allotments. - Figma will not receive proceeds from shares sold by existing stockholders. - The expected IPO price is $25–$28 per share. ## Listing and Underwriters - Figma has applied to list its Class A common stock on the New York Stock Exchange under the ticker symbol **FIG**. - Morgan Stanley, Goldman Sachs, Allen & Company, and J.P. Morgan are joint lead book-running managers. - BofA Securities, Wells Fargo Securities, and RBC Capital Markets are book-running managers. - William Blair and Wolfe | Nomura Alliance are serving as co-managers. ## Regulatory Status - Figma’s registration statement has been filed with the SEC but is not yet effective. - The shares cannot be sold, and purchase offers cannot be accepted, until the registration process is complete. - The offering will be made only through a prospectus. ## Figma’s Business - Founded in 2012, Figma describes itself as a collaborative platform for digital product development. - It has expanded from a design tool into a connected, AI-powered platform supporting ideation, design, development, and product delivery. Figma’s announcement marks a formal step toward becoming a public company, but the IPO remains subject to SEC approval and final offering conditions.

Read original

AI Figma Design Systems Prototyping+2

figmaJul 20, 20252 min readCurated summary

Mike Krieger and Luis von Ahn join Figma’s Board of Directors | Figma Blog

Figma announced that Mike Krieger, Anthropic’s Chief Product Officer and Instagram co-founder, and Luis von Ahn, Duolingo’s co-founder and CEO, have joined its Board of Directors. Both bring experience scaling widely used products and championing strong design, product quality, and innovation. Figma expects their perspectives to support its evolution into an AI-powered platform for collaborative product development. ## Mike Krieger’s Product and Infrastructure Experience - Krieger oversees product engineering, management, and design at Anthropic. - As Instagram’s co-founder and CTO, he scaled its infrastructure from a few million users to more than 1 billion monthly active users. - He later co-founded Artifact, a personalized news app acquired by Yahoo. - Figma CEO Dylan Field praised Krieger’s combination of broad product vision and meticulous execution. - Krieger was also an early angel investor in Figma. ## Luis von Ahn’s Design and Growth Expertise - Von Ahn co-founded Duolingo in 2011 and has led its growth to more than 100 million monthly active users. - Duolingo provides language education in over 40 languages and is known for its approachable, engaging design. - Before Duolingo, von Ahn co-created CAPTCHA and reCAPTCHA, including a crowdsourced approach to internet security. - Field highlighted von Ahn’s close collaboration with Duolingo’s design team and his advocacy for design despite his computer science background. ## Figma’s Expanding Board - Krieger and von Ahn join existing directors including Dylan Field, Mamoon Hamid, Kelly Kramer, John Lilly, Bill McDermott, Andrew Reed, Danny Rimer, and Lynn Vojvodich Radakovich. - The appointments add expertise in AI, consumer products, education, infrastructure, enterprise software, and design-led growth. - Figma describes itself as an AI-powered platform supporting the full process from ideation and design through development and shipping. Figma’s board additions reinforce its focus on product craft, design, AI, and large-scale user growth as it expands beyond design tooling into a broader product-development platform.

Read original

AI Figma Product Design Anthropic+4

googleJul 16, 2025Original article

Measuring heart rate with consumer ultra-wideband radar (opens in new tab)

Google Research has demonstrated that ultra-wideband (UWB) radar technology, which is already integrated into many modern smartphones for tasks like precise location and vehicle unlocking, can be repurposed for contactless heart rate monitoring. By employing a transfer learning approach, researchers successfully applied models trained on large datasets from Frequency Modulated Continuous Wave (FMCW) radar to the newer UWB systems. This development suggests that everyday consumer electronics could soon provide accurate vital sign measurements without the need for additional specialized sensors or physical contact. ## Leveraging Existing Consumer Hardware While Google previously used Soli radar (FMCW) for sleep sensing in the Nest Hub, UWB technology represents a more widely available hardware platform in the mobile market. * UWB is currently used primarily for non-radar applications like digital car keys and item tracking (e.g., Apple AirTags). * The technology is increasingly standard in high-end mobile phones, providing a ready-made infrastructure for health sensing. * Utilizing existing UWB chips eliminates the need for manufacturers to add dedicated medical sensors to devices. ## Overcoming Signal Interference in Vital Sensing The primary challenge in radar-based heart rate monitoring is that the micro-movements of the chest wall caused by a heartbeat are significantly smaller than movements caused by breathing or general body shifts. * The system utilizes three-dimensional spatial resolution to create a "measurement zone" focused specifically on the user's torso. * High temporal resolution, sampling at speeds up to 200Hz, allows the radar to capture the rapid, subtle pulses of a heartbeat. * By isolating reflections from the chest area, the radar can ignore stationary background objects and external movements that would otherwise corrupt the data. ## Cross-Radar Transfer Learning Because the researchers possessed extensive datasets for FMCW radar but very limited data for UWB, they developed a method to transfer learned features between different radar types despite their different physical principles. * FMCW radar transmits continuous sinusoidal waves, whereas UWB radar transmits extremely short pulses (picoseconds to nanoseconds). * The study used a large 980-hour FMCW dataset to "teach" the model the characteristics of human vitals. * This pre-trained knowledge was then applied to a smaller 37.3-hour UWB dataset, proving that heart rate features are consistent enough across hardware types for effective transfer learning. ## A Novel Spatio-Temporal Deep Learning Model The researchers designed a custom neural network architecture to process the complex multidimensional data generated by radar sensors. * The framework uses a 2D ResNet to analyze the input data across two axes: time and spatial measurements. * Following the initial analysis, the model uses average pooling to collapse the spatial dimension, focusing purely on the temporal signal. * A 1D ResNet then identifies long-range periodic patterns to estimate the heart rate. * The model achieved a mean absolute error (MAE) of 0.85 beats per minute (bpm), which is a 50% reduction in error compared to previous state-of-the-art methods. This research indicates that high-precision health monitoring can be integrated into the mobile devices users already carry. By transforming smartphones into passive health sensors, UWB technology could allow for continuous heart rate tracking during routine activities, such as sitting at a desk or holding a phone in one's lap.

AI Machine Learning Deep Learning Transfer Learning+5

lineJul 16, 2025Original article

LY's Tech Conference, ' (opens in new tab)

LY Corporation’s Tech-Verse 2025 conference highlighted the company's strategic pivot toward becoming an AI-centric organization through the "Catalyst One Platform" initiative. By integrating the disparate infrastructures of LINE and Yahoo! JAPAN into a unified private cloud, the company aims to achieve massive cost efficiencies while accelerating the deployment of AI agents across its entire service ecosystem. This transformation focuses on empowering engineers with AI-driven development tools to foster rapid innovation and deliver a seamless, "WOW" experience for global users. ### Infrastructure Integration and the Catalyst One Platform To address the redundancies following the merger of LINE and Yahoo! JAPAN, LY Corporation is consolidating its technical foundations into a single internal ecosystem known as the Catalyst One Platform. * **Private Cloud Advantage:** The company maintains its own private cloud to achieve a four-fold cost reduction compared to public cloud alternatives, managed by a lean team of 700 people supporting 500,000 servers. * **Unified Architecture:** The integration spans several layers, including Infrastructure (Project "DC-Hub"), Cloud (Project "Flava"), and specialized Data and AI platforms. * **Next-Generation Cloud "Flava":** This platform integrates existing services to enhance VM specifications, VPC networking, and high-performance object storage (Ceph and Dragon). * **Information Security:** A dedicated "SafeOps" framework is being implemented to provide governance and security across all integrated services, ensuring a safer environment for user data. ### AI Strategy and Service Agentization A core pillar of LY’s strategy is the "AI Agentization" of all its services, moving beyond simple features to proactive, personalized assistance. * **Scaling GenAI:** Generative AI has already been integrated into 44 different services within the group. * **Personalized Agents:** The company is developing the capacity to generate millions of specialized agents that can be linked together to support the unique needs of individual users. * **Agent Ecosystem:** The goal is to move from a standard platform model to one where every user interaction is mediated by an intelligent agent. ### AI-Driven Development Transformation Beyond user-facing services, LY is fundamentally changing how its engineers work by deploying internal AI development solutions to all staff starting in July. * **Code and Test Automation:** Proof of Concept (PoC) results showed a 96% accuracy rate for "Code Assist" and a 97% reduction in time for "Auto Test" procedures. * **RAG Integration:** The system utilizes Retrieval-Augmented Generation (RAG) to leverage internal company knowledge and guidelines, ensuring high-quality, context-aware development support. * **Efficiency Gains:** By automating repetitive tasks, the company intends for engineers to shift their focus from maintenance to creative service improvement and innovation. The successful integration of these platforms and the aggressive adoption of AI-driven development tools suggest that LY Corporation is positioning itself to be a leader in the "AI-agent" era. For technical organizations, LY's model serves as a case study in how large-scale mergers can leverage private cloud infrastructure to fund and accelerate a company-wide AI transition.

AI AI Agents Generative AI RAG+4

figmaJul 14, 20252 min readCurated summary

Why development leaders are investing in design | Figma Blog

Design has become a strategic driver of software success, not merely a visual layer added at the end of development. An IDC study of more than 500 development leaders found that teams prioritizing design report stronger business outcomes, faster delivery, and better collaboration. The research also shows that human design expertise is increasingly important as AI accelerates product creation. ## The Business Case for Design - 75% of development leaders consider design “very” or “extremely important” to modern software development. - Leaders who rated design as “extremely important” were five times more likely to say their last project far exceeded expectations. - The main business benefits associated with design investment were: - Improved customer retention - Higher customer engagement - Increased product innovation - Design now influences customer satisfaction, employee productivity, accessibility, and data-driven decision-making. ## Designer–Developer Collaboration - 52% of leaders said their organizations have increased software output over the past two years. - Collaboration between design and development from the beginning of a project helps teams: - Align on scope and trade-offs earlier - Reduce rework and friction - Iterate in real time - Improve morale and time to market - Among respondents, stronger design collaboration led to: - More innovation for 54% - Better customer experiences for 47% - Faster time to market for 43% - Effective collaboration was described as synchronized, structured, and mutually supportive rather than divided into isolated handoffs. ## Design Expertise in AI-Driven Development - AI makes it easier to create initial concepts, but generated outputs still require substantial human refinement. - Designers provide essential oversight in: - Quality control - User experience - Brand consistency - Cross-functional review - 80% of development leaders said design has become more important to the success of AI-powered products than it was two years ago. - The strongest teams treat AI output as a starting point, not a finished product. Human judgment is needed to turn fast-generated concepts into usable, user-centered experiences. Organizations seeking an advantage should treat design as a business priority, integrate designers and developers throughout the development process, establish scalable design practices, and preserve human oversight of AI-generated work.

Read original

AI User Experience Product Development Software Engineering+3

googleJul 9, 2025Original article

Graph foundation models for relational data (opens in new tab)

Google researchers have introduced Graph Foundation Models (GFMs) as a solution to the limitations of traditional tabular machine learning, which often ignores the rich connectivity of relational databases. By representing tables as interconnected graphs where rows are nodes and foreign keys are edges, this approach enables a single model to generalize across entirely different schemas and feature sets. This shift allows for transferable graph representations that can perform inference on unseen tasks without the costly need for domain-specific retraining. ### Transforming Relational Schemas into Graphs The core methodology involves a scalable data preparation step that converts standard relational database structures into a single heterogeneous graph. This process preserves the underlying logic of the data while making it compatible with graph-based learning: * **Node Mapping:** Each unique table is treated as a node type, and every individual row within that table is converted into a specific node. * **Edge Creation:** Foreign key relationships are transformed into typed edges that connect nodes across different tables. * **Feature Integration:** Standard columns containing numerical or categorical data are converted into node features, while temporal data can be preserved as features on either nodes or edges. ### Overcoming the Generalization Gap A primary hurdle in developing GFMs is the lack of a universal tokenization method, unlike the word pieces used in language models or patches used in vision models. Traditional Graph Neural Networks (GNNs) are typically locked to the specific graph they were trained on, but GFMs solve this through several technical innovations: * **Schema Agnosticism:** The model avoids hard-coded embedding tables for specific node types, allowing it to interpret database schemas it has never encountered during training. * **Feature Interaction Learning:** Instead of training on "absolute" features (like specific price distributions), the model captures how different features interact with one another across diverse tasks. * **Generalizable Encoders:** The architecture uses transferable methods to derive fixed-size representations for nodes, whether they contain three continuous float features or dozens of categorical values. ### Scaling and Real-World Application To handle the requirements of enterprise-level data, the GFM framework is built to operate on a massive scale using Google’s specialized infrastructure: * **Massive Throughput:** The system utilizes JAX and TPU infrastructure to process graphs containing billions of nodes and edges. * **Internal Validation:** The model has been tested on complex internal Google tasks, such as spam detection in advertisements, which requires analyzing dozens of interconnected relational tables simultaneously. * **Performance Benefits:** By considering the connections between rows—a factor traditional tabular baselines like decision trees often ignore—the GFM provides superior downstream performance in high-stakes prediction services. Transitioning from domain-specific models to Graph Foundation Models allows organizations to leverage relational data more holistically. By focusing on the connectivity of data rather than just isolated table features, GFMs provide a path toward a single, generalist model capable of handling diverse enterprise tasks.

AI Transformer Jax Graph Learning+5