Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

Figure

AI overview

Sign in and the AI will write an overview from our coverage.

Headlines · 9

  1. Humanoid Robots Shift From Stage Shows to Industrial Deployment

    TMTPost reports that Unitree Robotics shipped more than 5,500 pure-biped humanoids in 2025 and reported 868 million yuan in humanoid robot revenue, while the rental price for humanoid performances fell sharply as supply increased. Citing Counterpoint Research, the report says exhibitions, entertainment and data collection accounted for nearly 80% of first-half 2026 shipments, while factory and warehouse deployments remained below 20%. It argues that reliability, total cost of ownership, maintenance and recurring purchases still separate successful demonstrations from sustained commercial operation.

    钛媒体 · 🔥 0
  2. Sim-to-Real Gaps May Help Explain Uneven AI Progress

    An analysis argues that the sim-to-real gap may help explain why AI has made rapid progress in programming and mathematics while its real-world impact in many professions remains modest. Programming and mathematical reasoning can be replicated and verified in digital environments, whereas many white-collar professions involve people and physical environments that are difficult to simulate. The analysis uses robotics examples to illustrate the challenge. It also cites GDPval, where the best score reportedly rose from 47.6% for Claude Opus 4.1 when the benchmark was released to more than 85% for current leading models.

    Understanding AI · 🔥 0
  3. Humanoid robot makers converge on a 2027 proving ground

    TMTPost reports that multiple humanoid robot companies have placed major delivery, deployment, compute, demonstration, and standards milestones in 2027. Reported examples include Tesla targeting external Optimus delivery by the end of 2027, XPeng planning IRON launch and delivery that year, OpenAI promising a robot demonstration, and Figure expecting its 100,000-GPU cluster to come online in the second half. The analysis argues that inconsistent definitions of production, autonomy, and success, combined with limited third-party validation, make these timelines hard to compare.

    钛媒体 · 🔥 0
  4. OpenAI’s robot hiring signals a hardware push beyond “just the brain”

    OpenAI’s careers page reportedly listed 27 robotics-related positions in San Francisco in October 2026, including four dedicated to actuator design and one to actuator technical program management. TMTPost also reported that OpenAI OpCo LLC leased a 202,000-square-foot Richmond warehouse with power capacity exceeding 14,000 amperes, though the report does not establish how the site will be used. The listings concern open roles rather than 27 completed hires, and no production target or shipping schedule was disclosed.

    钛媒体 · 🔥 0
  5. Robots take sorting jobs, none is a full humanoid

    TMTPost reports from a China Post logistics center in Shenzhen's Longgang district that the robots deployed to parcel sorting in 2026 are not complete humanoids: some have a fixed lower body at a workstation with an upper body and dexterous hand or gripper, while others are arm-and-gripper rigs fixed at sorting points. The article argues sorting became embodied AI's early deployment site because it combines orders, repeat purchases, real-world data and a recirculation loop that tolerates mistakes. It relays vendor claims: Robot Era delivers a semi-humanoid body with its self-developed XHAND1 five-finger dexterous hand and says mature units run more than 10 effective hours a day; X Square Robot says an August 12 livestream with random parcels and no human backup reached 1,816 parcels per hour at over 98% success, and says its robots are already sorting at SF Express. The article also says Jijia Shijie signed a logistics framework agreement with Wuhan Post in August, Xinghaitu demonstrated a front warehouse with JD and signed a three-year cooperation, Daka Robotics ran a multi-robot picking livestream in a Tianjin warehouse, and AgiBot's G2 Max was deployed in JD's Zhilang warehouse.

    钛媒体 · 🔥 0
  6. Robots learning from human video: Robot Era opens VPP2 weights, Dyna-2 scale test

    On October 10, Robot Era released two sets of video model weights for VPP2. According to an article by Embodied Observer published via TMTPost, VPP2 splits robot skill learning into three stages — predicting how a task unfolds, compressing it into a visual plan, and learning robot actions — and only the first two were opened. The piece reviews several routes for putting human video into robot training: Dyna-2 compared pretraining on 1,000 to 1,000,000 hours of first-person video, and on a key-turning task only the 1,000,000-hour group succeeded, 9 times out of 10 tries; Nvidia's DreamZero raised average task progress from 38.3% to 54.3% after adding 12 minutes of action-label-free human video; τ0-WM used about 27,300 hours of heterogeneous data. The article also notes a gap between predicted video and executable action: DreamZero's oven task failed after the predicted plan skipped opening the door, and in RoboWM-Bench simulation, Wan 2.6 videos converted into actions succeeded 50% of the time at closing a drawer, 20% at picking an object and 0% at placing it in the drawer.

    钛媒体 · 🔥 0
  7. AI feeding: embodied-data collection grows into an industry chain

    An article published with authorization by TMTPost says embodied AI lacks physical-operation data and that a chain spanning collection, labeling and trading has formed. It cites Mifeng Technology CEO Yao Maoqing at the 2026 World AI Conference saying a model would need on the order of 100 million hours of real operation data to be as plug-and-play as GPT-3.5, while only hundreds of thousands of valid hours exist globally. The article also says 15 data-only service providers raised about 4.47 billion yuan in 2026 and that crowdsourced collection platforms are spreading, alongside low pay, subcontracting and consent concerns.

    钛媒体 · 🔥 0
  8. Physical AI funding tops $48B in 2026, beating all of last year

    Citing PitchBook data, the Financial Times reports that robotics and physical AI companies had raised nearly $48 billion globally by October 8, 2026 — roughly 1.7 times the whole of last year. Capital is concentrating in leaders: Anduril raised $5 billion in May at a $61 billion valuation, NEURA Robotics announced a C round of up to $1.4 billion in June, Skild AI raised $1.4 billion led by SoftBank in January, and Physical Intelligence raised $1 billion in the second quarter. Robot training data is also becoming its own industry: Figure's Index program has collected over 16 million videos and paid out $15 million, Dyna Robotics says its Dyna-2 model, pretrained on more than 1 million hours of human video, lifted average performance across 14 real tasks from 20% to 53%, and Skild AI reported annual recurring revenue above $100 million about 10 months after its first commercial deployment.

    量子位(原生 RSS) · 🔥 0
  9. Hark launches Hark Pro, a free proactive AI assistant for everyday tasks

    Hark, the startup founded by Brett Adcock, launched Hark Pro on Oct. 6 — a personal AI assistant available on web, iOS and Android, free at the standard tier with paid plans at $20/month for twice the usage and $100/month for ten times. It is built on a model trained specifically for computer use, and its cloud computer Handoff is said to operate as many as six browsers at once and log into sites on a user's behalf to order groceries, pay bills, book rides or reserve parking, with Action Buttons for one-tap approvals. Hark says credentials and payment details sit in an encrypted "Secured by Hark" vault that even the company cannot see, and says it will not sell user data to advertisers; first hardware is promised for 2027 with AT&T cellular service.

    TechRepublic · 🔥 0

Experience and discussion from the community

Share my Figure experienceAsk about Figure

phone-harness: Let your agent use your phone.

What problem it solves Coding agents need a way to interact with a real phone without requiring device modification or a native development environment. phone-harness gives agents a shared interface for reading screens and performing actions on iPhone, Android, and cloud phones. Key features Real-device phone control — Agents can scroll, tap, type, and read the screen on a real iPhone or Android w…

github.comhttps://github.com/phone-harness/phone-harness

itsaplan: Open-source project management and issue tracking where people and AI agents ship together.

What problem it solves Teams need an issue tracker they can use immediately and gradually extend to AI-assisted work. The project addresses this by treating agents as project members, supporting both scheduled and multi-agent workflows, and allowing the application to remain self-hosted on the team’s own server and database. Key features AI agents as project members — Agents can have their own rol…

github.comhttps://github.com/croffasia/itsaplan

FreeVideo: A local inference engine for generating MiniMax H3 videos on consumer GPUs.

What problem it solves FreeVideo enables local video generation on consumer hardware while coordinating VRAM, system memory, disk, weight placement, compute precision, and attention kernels. It provides a ComfyUI workflow and launchers that simplify setup across Windows, macOS, and Linux. Key features Hardware-adaptive execution — The engine selects GPU-specific weight formats and probes available…

github.comhttps://github.com/FlashML-org/FreeVideo

olake-go: An open-source data ingestion engine for replicating databases, S3, and Kafka into Apache Iceberg or plain Parquet.

What problem it solves Organizations need to move and continuously replicate data from transactional databases, event streams, and object storage into open lakehouse formats without depending on proprietary ingestion stacks. OLake Go addresses this with replication workflows, CDC, schema handling, and deployment options that avoid separate Spark, Flink, Kafka, or Debezium infrastructure. Key featu…

github.comhttps://github.com/datazip-inc/olake-go

autoclip: An open-source AI desktop tool that turns long videos into short clips, subtitles, covers, and publishing copy.

What problem it solves Long-form podcasts, interviews, and courses require extensive manual work to identify highlights, reframe footage, add subtitles, prepare covers, and write publishing copy. This project automates those steps from a source link while retaining local editing and rendering for users who need to adjust the result. Key features Platform-aware clip generation — Choose a target pla…

github.comhttps://github.com/zhouxiaoka/autoclip

nginx: An open-source web server, load balancer, reverse proxy, API gateway, and content cache.

What problem it solves NGINX provides software for serving web traffic and handling roles including load balancing, reverse proxying, API gateway functions, and content caching. Its modular architecture, text-based configuration, and process-based runtime give users multiple ways to configure and operate those capabilities. Key features Modular functionality — NGINX extends its core functionality …

github.comhttps://github.com/nginx/nginx

ai-agent-book: An open-source Chinese book that explains AI Agents from design principles to engineering practice through the formula Agent = LLM + context + tools.

What problem it solves Readers need a structured way to understand AI Agent design principles and apply them through engineering practice. The book addresses this by organizing the material into 10 chapters and pairing the text with 109 companion experiments covering topics such as context engineering, memory, tools, evaluation, post-training, continuous evolution, and multi-agent collaboration. K…

github.comhttps://github.com/bojieli/ai-agent-book

php-src: PHP is a general-purpose scripting language especially suited to web development.

What problem it solves Developers need a practical path to build, test, install, and extend the PHP interpreter from source. The repository also needs a clear contribution process for proposing new features, fixing bugs, and testing specific areas of the codebase. Key features Build from source — The repository provides a source-build workflow that generates a configure script, configures a debug …

github.comhttps://github.com/php/php-src

storybook: A frontend workshop for building UI components and pages in isolation.

What problem it solves UI components and pages need to be developed, tested, and documented independently from a complete application. Storybook provides an isolated workshop and an extensible API for these activities. Key features Isolated component workshop — Build and inspect UI components and pages in isolation, supporting focused work across development, testing, and documentation. Source: RE…

github.comhttps://github.com/storybookjs/storybook

sub2api: An AI API gateway platform for subscription quota distribution.

What problem it solves Managing subscriptions across multiple AI providers can require separate access paths and fragmented quota handling. This project provides a centralized gateway for distributing subscription quotas, supporting shared access and compatible use with native AI tools. Key features Subscription quota distribution — The gateway is designed to distribute and manage AI subscription …

github.comhttps://github.com/Wei-Shaw/sub2api

claude-code-security-review: An AI-powered security review GitHub Action using Claude to analyze code changes for security vulnerabilities.

What problem it solves Pull request changes can contain security vulnerabilities that require understanding their code context and intent. The action helps identify potential issues, filter false positives, and communicate findings directly in pull requests. Key features Context-aware security analysis — Claude examines changed code in context to understand its purpose and potential security impli…

github.comhttps://github.com/anthropics/claude-code-security-review

arnis: Creates detailed game worlds reflecting real-world geography, topography, and architecture.

What problem it solves Building representative game worlds from real locations requires geographic, elevation, and architectural data. Arnis addresses this by processing large-scale geospatial information to generate terrain and structures that reflect the selected area. Key features Multiple game editions — Generated worlds can target Minecraft Java Edition, Bedrock Edition, Luanti, and Hytale. S…

github.comhttps://github.com/louis-e/arnis

filebrowser: A self-hosted web-based file manager for accessing and managing files from the web.

What problem it solves People need a simple way to access and manage files through a web interface. FileBrowser Quantum addresses this with source configuration, user access controls, sharing, search, previews, editing, and a responsive interface. Key features Multiple source configuration — Configure multiple file sources with include and exclude rules for browsing and managing different location…

github.comhttps://github.com/gtsteffaniak/filebrowser

karakeep: A self-hostable bookmark-everything app for links, notes and images, with AI-based automatic tagging and full text search.

What problem it solves People who save articles, tools, notes, images and PDFs need a self-hosted place to organize and revisit that content. Karakeep addresses this with link previews, lists, collaboration, automatic tagging and summarization, full text and semantic search, and content extraction tools. Key features Save and organize varied content — Store links, notes, images and PDFs, automatic…

github.comhttps://github.com/karakeep-app/karakeep

Switchyard: An open-source library that helps an AI agent choose which model handles each request.

What problem it solves AI agents need to choose among models while balancing task success, cost, and latency for their workloads. Applications also need flexibility in where routing runs and who controls model clients, credentials, retries, and service operation. Key features Multiple routing algorithms — Routing can be configured with Auto, Task, Execution, or Composite strategies, including mode…

github.comhttps://github.com/NVIDIA-NeMo/Switchyard

netpulse: A self-hosted PWA dashboard that shows the live pulse of your home network, running from a single static Go binary with nothing leaving your LAN.

What problem it solves Home networks built on OpenWrt/GL.iNet routers are spread across many devices, and there is no single place to see fleet status, per-router health, connected devices, WireGuard peers, AdGuard Home statistics or alerts as they happen. Operators also need topology and health scoring without handing their network state to a third-party cloud service. NetPulse addresses this by …

github.comhttps://github.com/gnacho/netpulse

agenticos: Open-source, self-hosted AI agent platform for your company's AI teammates.

What problem it solves Most agent frameworks require code changes, reviews, and releases to modify an agent's behavior, which puts the person who understands the desired outcome behind an engineering process. Companies also need to share agents across teams while retaining control over access, spending, approvals, and recorded activity. Key features Browser-based agent building — Teams can configu…

github.comhttps://github.com/vstorm-co/agenticos

paper-orchestra: A multi-agent framework for automated AI research paper writing.

What problem it solves Researchers often need to turn unconstrained ideas and experimental logs into complete research manuscripts with literature synthesis, structured sections, refinements, and figures. PaperOrchestra addresses this gap by coordinating specialized agents across the paper-writing workflow and producing LaTeX manuscripts that follow target conference templates. Key features Multi-…

github.comhttps://github.com/google-research/paper-orchestra

erigon: An Ethereum execution layer with an embeddable consensus layer.

What problem it solves Ethereum node operators need an execution client that can integrate consensus functionality while supporting bulk block processing, historical queries, validator workloads, and flexible deployment configurations. Repository contributors also need guidance for building, testing, configuring, and extending the implementation. Key features Embedded and external consensus layers…

github.comhttps://github.com/erigontech/erigon

ratatui: A Rust crate for cooking up terminal user interfaces (TUIs).

What problem it solves Building text-based user interfaces for command-line applications, dashboards, and other interactive console programs requires a way to compose and render terminal interfaces. Ratatui addresses this problem with a simple and flexible API for creating TUIs, supported by templates, examples, and documentation. Key features Flexible terminal user interfaces — Create text-based …

github.comhttps://github.com/ratatui/ratatui