Computer Vision AI Startup Jobs

Data infrastructure company providing high-quality training data for AI applications, recently partnered with Meta.

Vast is a US-based commercial space company building human-rated artificial-gravity orbital space stations, starting with its single-module Haven-1 station targeted as the world's first commercial space station. Founded by Jed McCaleb and headquartered in Long Beach, California ('Space Beach'), the company designs and manufactures its spacecraft hardware in-house. It is racing to provide a commercial successor to the International Space Station.

Helsing is a European defense AI company building autonomy software and hardware for democratic nations. It develops real-time battlefield decision software, AI-enabled drones, and sensor fusion systems for military applications across NATO allies. Founded in 2021, it has grown to become one of Europe's most valuable defense tech startups with an $18B valuation.

Autonomous drone manufacturer building AI-powered drones for defense, public safety, and enterprise inspection.

Figure develops general-purpose humanoid robots and embodied AI systems for industrial and logistics environments, combining robotics hardware with AI control systems.

Nuro develops autonomous driving technology and licenses its Nuro Driver platform to automotive and mobility partners for commercial deployment.

Atoms is Travis Kalanick's physical-AI and industrial robotics holding company, rebranded in March 2026 from City Storage Systems. It operates business units across food (CloudKitchens, Otter, Lab37, Picnic), autonomous mining (built on acquired robotics firm Pronto), and transport, building a shared robotics 'wheelbase' for specialized industrial machines rather than humanoids. The company runs operations in 110+ cities and raised $1.7B led by a16z in July 2026, with Ben Horowitz joining the board.

Apex is a defense-grade satellite bus manufacturer producing configurable, productized spacecraft platforms (Aries, Nova, Comet) for LEO, MEO, and GEO orbits. The Los Angeles-based company serves commercial and sovereign government customers at scale with peak annual capacity exceeding 200 units per year. Apex launched Project Shadow, the first commercially-led on-orbit space-based interceptor demonstration.

1X Technologies builds general-purpose humanoid robots designed for home and industrial use. Their flagship products include EVE for industrial facilities and NEO for the home. Backed by OpenAI.

Waabi develops AI-first autonomous trucking software and simulation systems to accelerate validation and deployment of self-driving transport technology.

Building general-purpose robotic intelligence and foundation models for adaptable robots across environments and tasks.

XYZ Reality builds augmented-reality construction delivery technology that overlays digital models on site to improve layout precision, quality assurance, and project controls.

RobCo is a Munich-founded industrial robotics startup providing modular, plug-and-play robotic hardware kits and AI-powered automation software for small and mid-sized manufacturers. Its RobCo Studio software enables no-code programming of robots for tasks like machine tending, palletizing, welding, and dispensing. Backed by Lightspeed and Sequoia, it is scaling US operations with offices in San Francisco and Austin.

Standard Bots builds AI-native industrial robot arms and the software stack that runs them, targeting manufacturers who want automation without dedicated robotics engineers. Its RO1 cobot combines a no-code/AI control interface with US-based manufacturing in Glen Cove, NY. The company hit a $1B valuation with a $200M Series C in June 2026.

Encord is a data development platform for AI that helps teams curate, manage, and label training data at scale. The company serves 300+ AI teams including Toyota and Skydio, processing over 5PB of data.

Runway is an applied AI research company and pioneer of generative media tools, building AI-powered software for video, image, and multimedia content generation. Best known for text-to-video and image-to-video models used by filmmakers and content creators.

NODA AI is an AI-native defense company building autonomy orchestration software that enables command and control of mixed autonomous and manned assets across multi-domain military operations. Its platform serves as a vendor-agnostic reasoning layer coordinating heterogeneous autonomous systems for the Department of Defense and allied defense customers.

Array Labs builds high-power, low-cost space radar systems and a formation-flying radar constellation for large-scale 3D Earth imaging and persistent monitoring.

Machina Labs develops AI-powered robotic manufacturing systems for sheet metal forming and industrial production, reducing the need for traditional tooling.

AI-powered fleet safety and driver analytics.

Overland AI develops autonomous ground vehicle software and systems for off-road and contested environments, with defense-focused robotics deployments.

Mythic builds power-efficient analog compute technology for AI inference, targeting edge-to-enterprise deployments where energy efficiency and throughput are critical.

BRINC Drones builds public-safety drone systems for 911 rapid response, indoor tactical operations, firefighting, and search and rescue, including the Lemur 2, Responder, and Guardian aircraft plus its LiveOps command-and-control software. Its platform integrates with Motorola Solutions' police radios, 911/CAD, and CommandCentral systems and is used by 900+ public-safety agencies and more than 20% of U.S. SWAT teams. Founded by CEO Blake Resnick and headquartered in Seattle, the company is scaling U.S. manufacturing with a new Queen Anne factory and HQ.

TerraFirma is an Austin-based, vertically integrated construction robotics company founded in 2024 by former SpaceX engineers Noah Schochet (CEO) and Noah McGuinness (CTO). It retrofits heavy equipment — excavators, dozers, loaders, rollers, skid steers — for AI-assisted remote and semi-autonomous operation, paired with AI construction planning software. Rather than only selling robots to contractors, TerraFirma takes on earthwork, demolition, site prep, and infrastructure projects itself for data centers, roads, substations, and energy sites.

Orchard builds AI-powered robotic systems that automate orchard harvesting with computer vision and advanced robotics.

Sunday is a robotics company building Memo, a household humanoid robot designed to assist families with everyday chores like dishes, laundry, and tidying. Founded by Stanford-trained roboticists Tony Zhao and Cheng Chi, the company uses data from thousands of real homes to train its robots. Sunday reached a $1.15 billion valuation with its Series B round and is preparing to deploy its first robots through a Founding Family Beta program.

Flock Safety builds AI-powered cameras and software that help law enforcement and communities solve and prevent crime. The platform uses machine learning and computer vision to provide investigative intelligence for cities, neighborhoods, and businesses.

Menlo Research is a Physical AI research lab that began with Jan, an open-source, privacy-first AI assistant that has surpassed 6 million downloads, and is now building robots and models to solve real-world physical tasks. The ~25-person team is distributed across Singapore, Vietnam, and the US. It is community-built with no disclosed institutional funding rounds.

Odyssey is an AI lab building interactive video and world models that generate real-time, explorable scenes for new media.

Monumental develops AI-powered autonomous robots for on-site masonry construction work, performing bricklaying with millimeter precision using computer vision to make construction software-defined and autonomous.

Roboflow provides end-to-end computer vision tools that enable developers and enterprises to build, deploy, and manage vision AI models. The platform supports the full lifecycle from data labeling and model training to inference deployment, serving over 250,000 developers and powering applications across industries including manufacturing, retail, and agriculture.

MarqVision builds AI-powered brand protection and IP operations software that helps global brands detect, prioritize, and enforce against counterfeit and piracy threats at scale.

Meshy builds generative AI tools for creating production-ready 3D models and textures from text or images for game, XR, and creative workflows.

Neuromorphic computing startup building brain-inspired analog hardware to achieve biology-scale energy efficiency for AI workloads, founded by former Databricks AI head Naveen Rao.

Generalist AI is building embodied foundation models for general-purpose robots, starting with a focus on dexterous manipulation, leveraging research from OpenAI, Boston Dynamics, and Google DeepMind to create end-to-end deep neural networks that map sensor data to real-time robotic actions.

Mecka (Mecka AI) is a New York–based startup building the data engine for Physical AI — it collects real human motion data from body sensors and iPhones to train robots and embodied AI. The company provides the data, evaluation, and deployment infrastructure that helps robots learn from real human activity and move from lab demos to reliable real-world work.

AI-enabled image editing app offering background removal, retouching, and batch editing services, processing 5 billion images annually and serving over 150 million downloads for e-commerce and social media content creators.

TwelveLabs builds multimodal foundation models that understand, index, and search video content for developers and enterprises.

Carbon Robotics develops AI-powered agricultural robotics, including its flagship LaserWeeder which uses computer vision, deep learning, and CO2 lasers to identify and eliminate weeds with millimeter accuracy. The company also offers Carbon ATK, a tractor autonomy solution, and operates across 15 countries supporting over 100 crop types. Founded in 2018 and headquartered in Seattle.

Dexterity develops physical AI robotic systems for industrial automation, enabling warehouses and logistics operations to automate complex manipulation tasks.

OpenArt is a San Francisco-based AI image and video creation platform founded by ex-Googlers, serving roughly 8M monthly active users. Its flagship product, Director, generates complete videos up to about five minutes long from natural-language descriptions of story, characters, style, and pacing — an approach the company calls Vibe Directing. Users produce short films, trailers, micro-dramas, music videos, and UGC ads without editing software or technical expertise.

Quartermaster (legally Quartermaster AI Inc.) builds SmartMast™, a weather-hardened COTS sensor package — HD camera, software-defined radio, and onboard AI — that mounts on commercial and civilian vessels and turns them into nodes of a distributed maritime sensing network. The platform fuses video, radar, RF sensing, and AIS into near-real-time maritime domain awareness for coast guards, navies, insurers, energy operators, and researchers, positioning itself as a replacement for the spoofable, opt-in AIS standard. The company reports 600+ sensors deployed across 25+ countries and over 400,000 vessels identified outside of AIS.

Coframe builds an AI platform that continuously tests and optimizes website interfaces and marketing campaigns using generative and multimodal models. The platform autonomously improves text, visuals, and code, delivering up to 352% lift in click-through rates.

Tavus is the human computing company building emotionally intelligent AI humans (PALs - Personal Affective Links) that communicate through text, voice, and face-to-face video with agentic capabilities and real-time perception. Powered by proprietary models including Phoenix-4, Sparrow-1, and Raven-1, Tavus enables AI agents that can see, hear, understand emotions, and act autonomously. Over 100,000 developers and enterprises use Tavus technology for recruiting, sales, education, and customer service.

Bright Machines provides software-defined manufacturing and industrial automation systems that combine robotics, AI, and computer vision for electronics production.

Casa is a subscription-based home maintenance platform that pairs an AI-driven concierge with a network of vetted handymen to handle repairs, projects, and proactive upkeep for homeowners. It uses lidar scanning and AI to catalog a home's systems and schedule maintenance, sold as a monthly membership (~$200/mo). Founded by ex-Uber/CloudKitchens operators Michael York and Michael Mizrahi, it currently operates in the Bay Area and Los Angeles.

Gather AI builds autonomous drone systems for warehouse inventory monitoring. Its AI-powered drones and software platform enable logistics operators to scan facilities, count stock, and achieve inventory visibility without manual cycle counts. The company was founded in 2017 by Carnegie Mellon alumni and serves modern warehouses and distribution centers.

AI-powered baby monitoring company using computer vision to track infant sleep, health and development, trusted by over one million families with upcoming Parenting Intelligence System

Sunrise Robotics builds intelligent robotic cells for manufacturing, combining AI software, computer vision, and industrial automation to take repetitive work off factory floors.

Corvus Robotics builds fully autonomous warehouse inventory drones that scan, count, and track pallets without facility modifications or external infrastructure. Its Level 4 autonomous drones navigate aisles and read barcodes to automate inventory management for warehouses and manufacturing facilities. The company sells this inventory-and-inspection capability as a service to enterprise logistics operators.

Krea is a next-generation AI creative suite that provides a unified platform combining multiple generative AI models with a custom interface for image, video, and multimedia creation. It is designed to give designers and creatives intuitive, fine-grained control over AI-generated content. Customers include Pixar, LEGO, and Samsung.

Squint is an AI-powered manufacturing intelligence platform that helps industrial frontline operators reduce downtime and improve quality at scale. The platform delivers AR-guided procedures and knowledge-sharing tools to factory floors, serving customers like PepsiCo, Michelin, and Ford.

Until Labs develops reversible cryopreservation technology to pause biological time, starting with organ-scale preservation to extend the viability of donor organs for transplantation. The company is building medical infrastructure toward whole-body medical hibernation technology.

Pika develops consumer-first generative video tools that let users create and edit short-form videos from prompts, images, and clips through an intuitive AI-native workflow.

AI-powered design collaboration platform for industrial and product designers, enabling rapid concept sketching, rendering, and 3D visualization from rough sketches.

Ataraxis AI is a New York-based clinical AI research lab building multi-modal precision-medicine models that predict patient outcomes and treatment response. Its foundation models (Falcon, Kestrel) for digital pathology power products like Ataraxis Breast, an AI-native prognostic test for breast cancer. Founded in 2023 as an NYU spinout, it counts Yann LeCun among its founding advisors.

Chef Robotics builds AI-enabled robotic systems for food factories, prepared-meal production, and food assembly operations. Its robots automate repetitive plating, portioning, and packaging tasks in commercial food production, offered via a Robotics-as-a-Service model. The company has deployed systems that have produced tens of millions of meals.

Synthetic data generation platform for autonomous vehicle training and simulation.

Generative design software for fashion and consumer brands to rapidly create on-brand visual concepts.

World Labs is advancing spatial intelligence by building foundational AI models that can perceive, generate, and interact with 3D worlds. Founded by AI pioneer Fei-Fei Li, the company's flagship product Marble generates and edits persistent 3D environments from text, images, video, or 3D layouts. Backed by Andreessen Horowitz, Nvidia, AMD, and Autodesk, World Labs has raised over $1.2 billion to bring world models into creative and industrial workflows.

Alloy is a data platform for robotics that helps companies process, organize, and search through the massive volumes of sensor, camera, and telemetry data their robots generate. The Sydney-based startup enables natural language search across robot data and automated issue detection, reducing data processing time by up to 90%.

Doji is an AI fashion shopping app that creates a personalized avatar so consumers can try on real apparel products, explore looks, and share style ideas before purchasing.

AI video generation platform that creates lifelike animated characters by combining text, image, and audio inputs.

Helm.ai develops AI software for autonomous driving (AD/ADAS) and robotics automation, using an unsupervised 'Deep Teaching' approach that trains neural networks without human annotation or simulation. It partners with major automotive OEMs, and in October 2025 received an additional strategic investment from Honda to accelerate next-generation AD/ADAS development.

Computer vision company spun out of Intel, developing AI-powered smart vision systems for robotics, biometrics and access control markets using stereoscopic depth sensing technology

AMI Labs is a frontier AI research lab building world-model-based intelligence systems with planning, memory, and safety controls for real-world applications.

Apex.AI builds safety-certified, developer-friendly software for autonomous and software-defined vehicles, including its Apex.Grace (formerly Apex.OS) middleware derived from the open-source ROS 2 framework. Its products power autonomous mobility and robotic systems for automotive, trucking, and agricultural-machinery OEMs. The company is headquartered in Palo Alto, California, with offices in Germany, Sweden, Japan, and South Korea.

Hugging Face is the leading open-source AI platform for building, training, and deploying machine learning models. It hosts hundreds of thousands of models and datasets, serving as the central hub for the AI developer community.

Open-source multimodal data logging and visualization SDK for robotics and computer vision teams, enabling engineers to log, replay, and inspect spatial and temporal data.

The Bot Company is a San Francisco robotics startup founded in 2024 by former Cruise CEO Kyle Vogt, ex-Tesla AI lead Paril Jain, and ex-Cruise engineer Luke Holoubek. It is building an affordable, non-humanoid household robot that uses AI to pick up, organize, and put away items so people spend less time on chores. The company has raised over $550M pre-product, most recently at a valuation above $4B.

Frontier AI lab building reasoning models that let anyone describe a game or simulation in natural language and see it transformed into a working interactive 2D/3D scene within minutes.

Pangram Labs builds AI-generated content detection systems for text and images, claiming over 99% accuracy on AI-assisted and mixed human-AI writing. Its products serve higher-education institutions, publishers, media integrity organizations and businesses, and it publishes research on state-of-the-art detection techniques. Founded by Stanford AI graduates Max Spero and Bradley Emi, the team works in person in Brooklyn, New York.

Pantograph is a San Francisco-based public benefit corporation and applied research lab building generally intelligent, low-cost robots. It trains general models on internet-scale video and video games, then deploys them onto its own fleet of affordable, durable robots for large-scale real-world interaction data collection. The company owns its full stack — hardware, firmware, training infrastructure, and learning algorithms — and released its first video-pretrained model in July 2026.

Picsart is an AI-powered creative platform for photo and video editing with over 150 million monthly active users. It provides AI-driven tools for content creation, background removal, and design for both consumers and businesses.

Sable is a San Francisco-based startup building Aidan, an AI employee that joins live customer calls to run real-time product demos using computer use, vision, and voice inside a real browser environment. Founded in October 2025 by four Harvard-connected founders, it is already deployed at companies like Notion and Decagon. Sable raised a $45M round co-led by Sequoia Capital and 8VC in July 2026, with Shaun Maguire and Joe Lonsdale joining its board.

AI-powered infrastructure intelligence platform for transportation infrastructure management using LiDAR sensors, cameras, and advanced AI algorithms.

Ecorobotix is a Swiss agricultural technology company that develops AI-powered precision spraying systems, including the ARA field sprayer, which uses computer vision and machine learning to detect individual plants and apply herbicides with up to 95% less chemical usage. Founded in 2014 and headquartered in Yverdon-les-Bains, Switzerland, the company has deployed approximately 1,000 ARA units across Europe and operates in over 20 countries.

GenPeach AI develops multimodal generative foundation models for human-centric image and video creation, powering digital humans, virtual influencers, and AI companions for creative industries. Founded in 2025 and based in Zurich, Switzerland (with team members in Warsaw), it is an early-stage startup backed by angels and founders from OpenAI, Meta AI, Google DeepMind, and Microsoft AI. Per PitchBook/Nordic9 it has raised roughly $1.5M in pre-seed funding.

Loti AI protects major celebrities, public figures, and corporate IP from online threats specializing in deepfake and impersonation detection with advanced facial and voice recognition software that analyzes real photos, video and audio then scans the internet for infringing media with 95% takedown success rate.

An AI video security system that transforms existing IP cameras into intelligent agents for real-time detection and automated monitoring in healthcare and other sectors.

MatrixSpace develops AI-enabled 4D radar sensing and edge-to-cloud situational awareness software for defense, security, and critical infrastructure applications.

Sunflower Labs builds autonomous drone-based security systems that combine robotics, sensors, and software to monitor residential, commercial, and industrial properties.

Third Dimension AI is a 3D generative AI platform that creates high-fidelity, rendering engine-ready environments from text prompts and images. The platform serves professionals in gaming, film, autonomous vehicles, and defense.

Voliro develops autonomous aerial robotics for industrial inspections, combining a drone platform, tiltable payload system, and software for high-value asset inspection workflows.

Blackstar is an AI hardware startup building a new computing device and operating system aimed at fundamentally improving how humans interact with AI. Founded by former OpenAI Codex engineer Daniel Edrisian, the ~8-person team is split between San Francisco and Shenzhen and has not yet shipped a public product.

Citizen is an NYC-based consumer safety app that delivers real-time alerts about nearby incidents using 911 radio feeds, user-submitted video, and location data. Launched in 2016 as Vigilante and rebranded in 2017, it also sells Citizen Protect, a paid subscription offering 24/7 access to live safety agents. The company says over 9 million people use the service across major US cities.

OpenSpace builds AI-powered reality capture and analytics software that helps construction teams track project progress and reduce field risk.

Phota Labs is building personalized generative AI experiences for photography, focused on making high-quality visual creation and enhancement accessible to mainstream users.

Recraft is a text-to-image model and design platform for photorealistic assets, vector generation, custom brand styles, mockups, and collaborative creative workflows.

Shade is an AI-native unified file system for creative and post-production teams, combining cloud storage, real-time large-file streaming into editing tools, and natural-language search across footage and imagery. The platform consolidates the fragmented post-production stack (cloud storage, streamable file systems, media asset management, and review tools) into one system and has ingested 60M+ assets for brands, agencies, and sports organizations.

Wonder is the first AI-native design tool built on an infinite canvas that understands your components and design patterns, automatically suggesting the right elements to keep work consistent while accelerating creative exploration. The platform enables designers to explore layouts, UX flows, and build design systems with AI-powered assistance.

Wonder Studios is an AI-native film and content production studio harnessing generative AI to develop original films, music videos, and branded content. Founded in London with offices in New York and Los Angeles, the studio works with a community of over 200 AI-native filmmakers and has produced its first AI-generated anthology series.

Zanskar is an AI-powered geothermal exploration company that uses custom-built machine learning models to discover untapped conventional geothermal resources. The platform combines seismic characterization, directional drilling guidance, and reservoir modeling, and develops utility-scale geothermal power plants across the western United States.

Bunkerhill Health builds Carebricks, an agentic AI platform that lets hospitals and health systems build, deploy, and govern production-grade AI agents for clinical, operational, and administrative work — from cardiac imaging analysis and lung nodule detection to patient triage and insurance prior authorization. Customers include Cleveland Clinic, UTMB, Intermountain Health, and Mayo Clinic, with UTMB running 20+ agents on the platform. The YC W19 company raised a $25M Series B led by Khosla Ventures in July 2026, bringing total funding to $55M.

Building the world's most biomimetic musculoskeletal androids featuring over 200 degrees of freedom and proprietary artificial muscles. The Protoclone V1 can be manufactured for under $20,000.

Sudo Robotics is building general-purpose robots powered by a manipulation-centric foundation model, with its flagship #sudo R1 system integrating self-developed hardware and software. The system trains manipulation policies in simulation alone, using reactive observation-conditioned control at 15-25 Hz to achieve high reliability in object picking and grasping. The company targets physical-AI applications across manufacturing, logistics, agriculture, and eldercare.

Eyebot builds an AI-powered eye exam and vision-care platform with automated kiosks to expand access to eye care.

InspectMind AI (YC W24) builds an AI agent that reviews construction drawings and catches design issues before they become costly rework. The platform cross-checks architecture, engineering, and specifications to detect conflicts and compliance problems in minutes, helping teams avoid downstream errors.

Lazarus AI (Lazarus Enterprises, Inc.) is a Boston/Cambridge, MA-based enterprise AI company focused on document understanding and intelligent data processing. Its platform — OCR/Forms extraction, the RikAI reasoning engine, and a Vector Knowledge Graph — extracts and reasons over documents, images, and other unstructured data without training, serving insurance, public-sector, and healthcare customers. It received undisclosed strategic investments from WaveFx (Aug 2024) and QBE Ventures (Oct 2024).

Multimodal AI video clipping platform that automatically finds, edits, and posts short-form social content from long videos. Used by media companies like iHeart, marketing teams, and podcasters to 10x social impressions with autonomous video workflows, custom-trained agents, and multi-platform scheduling for TikTok, Instagram, YouTube Shorts, and X.

AI-powered IP protection platform that automates monitoring, takedown, and retroactive compensation for brand infringement across 7,800+ platforms. Cuts enforcement time by up to 90%.

Prometheus is a Physical AI company co-founded by Jeff Bezos and Vik Bajaj building an 'artificial general engineer' — AI that can design, experiment on, and manufacture complex physical products such as jet engines, drugs, and advanced materials. The company operates out of San Francisco, London, and Zurich, recruiting researchers from OpenAI, DeepMind, and Meta. It raised a $12B round in June 2026 at a ~$41B valuation, bringing total funding above $18B.

V7 provides a data engine for computer vision teams to manage datasets, label data, and train AI models at scale.

Backflip AI turns text prompts into 3D models and design-ready assets for physical products.

Flora is a node-based AI design tool and infinite canvas platform for creative professionals. The platform allows designers to build and automate creative workflows using a visual node orchestration layer, enabling creation of images, videos, and other media assets. Used by designers at Alibaba, Brex, Pentagram, and Lionsgate.

Onton is an AI-powered product search and discovery engine that uses neuro-symbolic, multimodal search to help shoppers find the right product. Originally focused on furniture (formerly Deft), the platform is expanding to apparel and consumer electronics. It aims to dramatically shorten shopping decision time compared to traditional e-commerce browsing.

Aether AI is a frontier AI research company building causal world models — systems that reason through causal mechanisms rather than statistical correlations, so machines can reason under interventions and act reliably in complex real-world environments. Founded by UC San Diego professor and causal-discovery researcher Biwei Huang, its initial commercial focus is Physical AI and robotics. The company raised a $20M seed round in June 2026 led by MPCi.

AI Clearing provides AI-powered construction progress monitoring and reporting software. Its platform leverages machine learning and drone-captured data to deliver automated digital field tracking, 3D site reports, and 4D geospatial analytics for energy, utilities, and infrastructure projects.

AirWorks provides geospatial AI and aerial mapping workflows to convert raw field data into utility and infrastructure-ready design deliverables.

Aleph Neuro is a San Francisco research lab building non-invasive brain interfaces, aiming to read and transmit "latents" — raw thoughts — toward direct brain-to-brain communication. In June 2026 it publicly launched with what it claims are the highest-resolution 3D images of the human brain captured from outside the skull, delivering MRI-level detail without surgery or bulky hardware. The system is built on Butterfly Network's Ultrasound-on-Chip technology via the Butterfly Embedded licensing and co-development program.

Alta is an AI-powered personal stylist app that organizes your wardrobe, creates outfit recommendations, and enables virtual try-ons using a digital avatar. Backed by Menlo Ventures and LVMH-backed Aglae Ventures, Alta leverages over a dozen proprietary multimodal AI models trained in-house on fashion data.

AI headshots at scale-turn selfies into studio-quality portraits for teams and individuals.

Armstrong Robotics builds general-purpose robots for commercial kitchens, starting with AI-powered dishwashing automation that handles messy, unpredictable environments with commercial-grade reliability. Its systems use robot arms with 3D perception trained on thousands of hours of real dishwashing data and integrate with existing commercial dish machines on a subscription model. The platform is designed to extend to other repetitive kitchen tasks beyond dishwashing.

Augrade is an AI-powered spatial computing platform that automates heavy industry design and engineering, generating manufacturing and construction-grade CAD models in minutes.

Axis Robotics builds a distributed data engine for physical AI, letting a global network of contributors generate robot training data through browser-based teleoperation and mobile collection tools. Its simulation-first stack combines large-scale pretraining data, hybrid augmentation, and corrective post-training feedback into a continuously improving loop for embodied AI models. The network claims 100,000+ contributors producing roughly 1,200 hours of simulation and 20,000 hours of real-world data monthly, with commercial partners including Booster Robotics and Geely Auto.

Black Forest Labs is a generative AI company that created FLUX, one of the most advanced open-source image generation models, and previously pioneered Latent Diffusion and Stable Diffusion. Founded by machine learning researchers Robin Rombach, Patrick Esser, and Andreas Blattmann, the company builds powerful creative AI models from labs in Freiburg, Germany and San Francisco. Valued at $3.25 billion after raising $300 million in Series B funding, Black Forest Labs is at the forefront of generative media technology.

Butlr is a privacy-first ambient sensing platform spun out of MIT Media Lab that uses thermal sensors and AI to detect human presence, occupancy, and posture without capturing personally identifiable information. The sensors serve enterprises across workplace optimization, senior living, and building energy management.

Cartwheel builds AI animation tools that turn text, motion, and video inputs into production-ready character animation for games, film, advertising, and social content.

Cerrion is a Zurich-based industrial-AI company that turns standard factory cameras into 24/7 AI video agents which detect process deviations, quality defects, and safety issues on production lines and intervene automatically. Founded in 2020 and backed by Y Combinator, its platform is deployed across 15 countries with customers including Unilever and Riedel.

ConeLabs provides AI-powered inspection solutions for construction and infrastructure. Using advanced drone technology and 3D modeling, the platform captures high-resolution images of structures, converts them into 3D models, and uses AI to detect, identify, and categorize defects across buildings, bridges, utilities, and transport infrastructure.

AI-powered visual content generation platform leveraging computer vision technology for image and video editing, offering background removal, image enhancement, and AI content generation tools developed by LibAI team.

Decart is an AI research lab building real-time generative and world models plus the inference stack that runs them. Its products include DOS (a hardware-optimized training/inference layer that lets models run across different AI chips), Lucy (live video transformation for virtual try-on and interactive media), and Oasis (an interactive world model for gaming, robotics, and physical AI). Founded in Israel with offices in Tel Aviv and San Francisco.

Experimental AI product lab offering tools for content generation and editing across various media, including image and video generation, music composition, and photo editing

Drafted is a Y Combinator-backed AI startup building multimodal generative models for residential architecture and spatial design. Its platform turns simple inputs (lot size, room count, target square footage) and sketches into custom floor plans, exterior elevations, and 3D models, exporting CAD/BIM-compatible files (Revit, IFC, DXF). Founded by former Atmos founder Nick Donahue, it hit 120k users generating 325k+ floor plans in its first month.

Flick is an AI-native filmmaking platform that gives professional creators cinematic control through an AI-native workflow built for storytelling, iteration, and artistic direction. Co-founded by award-winning filmmaker Zoey Zhang and founding Instagram engineer Ray Wang, the YC F25 startup raised a $6M seed round in May 2026 from True Ventures, GV, YC, Lightspeed, Formosa Capital, Pioneer Fund, Olive Tree Capital, and N1.

Gatsby is a San Francisco-based on-demand consumer platform for humanoid robot cleaning services, operating under parent entity West Egg Labs. Founded in January 2026 by Aron Frishberg, the company is robot-agnostic and builds the consumer distribution layer for humanoid robotics rather than competing on hardware. The service is live in San Francisco via an iOS app.

General Intuition is a public-benefit frontier research lab building foundation models for environments that require deep spatial and temporal reasoning. It starts with games as a rich training ground and aims to transfer those capabilities into the real world.

Builds fully autonomous, cabless electric freight haulers (Class 8 trucks) that use vision-based AI to move freight across warehouses, ports, and railyards at lower cost.

Instance (Instance Labs) automates evaluation of robot learning policies. Its product analyzes rollout video and returns success/failure verdicts with per-subtask breakdowns and evidence, replacing the manual work of watching rollouts, judging outcomes, and resetting test environments. Founded 2026 by MIT alums Claire Mao (CEO) and Lucy Cai (CTO), part of Y Combinator's S26 batch.

Instrumental builds an AI-powered manufacturing optimization platform for complex electronics, combining in-line camera hardware with computer-vision models that detect defects, surface root causes, and enforce quality controls across contract-manufacturing lines. Founded by two ex-Apple engineers, it is used by consumer-electronics and data-center hardware makers to improve yield and speed up failure analysis. In March 2026 it announced an investment from NVIDIA's NVentures tied to accelerating AI server and rack manufacturing.

Interhuman AI builds the social intelligence layer for AI systems, enabling them to read, interpret, and respond to human behavior in real time. Their technology combines body language, facial expression, and tone-of-voice analysis with contextual awareness to create more natural human-AI interactions.

AI-based platform that converts 2D images into production-ready 3D models, combining machine learning technology with expert 3D artists to deliver assets 10x faster for game studios and brands.

Lemon Slice is a research and product lab building real-time interactive AI avatars, using its own video-diffusion models to turn any single image into a character you can video-chat with live. Its Lemon Slice-2 model is offered to developers via API and embeddable web components for use cases like education, language learning, e-commerce, and corporate training. The company emerged from Y Combinator's W24 batch.

LGND is building 'ChatGPT for the Earth,' a geospatial AI infrastructure platform that makes the world's satellite and geographic data as accessible and searchable as text. Using transformer-based geographic embeddings, LGND enables teams to create, adapt, and scale geospatial datasets across time and geography with reduced technical overhead.

Creates generative 3D and video models that capture the real world and render cinematic scenes from text or images.

MicroAGI is a Berlin-based data research lab accelerating embodied AI by recording real-world human task execution through camera-wearing workers, generating structured egocentric video datasets for robotics and AI labs. The company runs active data-collection operations in NYC through its Shift app, targeting the fast-growing market for physical-world AI training data.

Independent research lab creating generative AI tools for image generation, known for producing high-quality artwork from text descriptions and serving over 21 million Discord community members.

Nanonets is a document AI and intelligent workflow automation platform that extracts structured data from invoices, purchase orders, contracts, and other business documents using no-code machine learning. The platform automates back-office workflows and integrates clean data into systems of record like SAP and Salesforce. The company raised a $29M Series B in March 2024 led by Accel, bringing total funding to ~$42M.

NEURA Robotics is a German company building cognitive and humanoid robots powered by embodied AI, including the 4NE-1 humanoid and the MiPA personal-assistant robot. Founded in 2019 and headquartered in Metzingen, Germany, it develops full-stack robotics that combine proprietary hardware, sensors, and AI software for industrial and service use. In June 2026 it announced a record Series C of up to $1.4B led by Tether, with Amazon, NVIDIA, and Qualcomm participating.

Noitom Robotics develops embodied intelligence technology for humanoid robots, providing high-fidelity motion capture, teleoperation systems, and multimodal data acquisition platforms used to train robot perception and locomotion. The company positions itself as a data infrastructure provider — 'a robot company that doesn't build robots' — supplying scalable training data and tooling to robot manufacturers and embodied AI model teams. It spun out of Noitom Technology, a global leader in professional motion capture.

NORI Robotics is a San Francisco-based startup building affordable bimanual home robots. Its flagship NORI L2 is a two-armed household robot starting at $1,288 that autonomously handles chores like loading the dishwasher, folding laundry, and cleaning, with a skill-sharing marketplace and remote-operation support. The robot is made in the US and slated to ship summer 2026.

AI-powered photo editing platform offering tools for background removal, image upscaling, and automatic photo retouching to enhance visual content for e-commerce, advertising, and social media.

Leading 3D scanning and capture platform for iOS, Android, and web with LiDAR technology, photogrammetry, and 360° capabilities for creating professional-grade 3D models and floor plans.

Prentis is an AI research lab building computer-use models that perceive screens and operate software directly across mobile, browser, and desktop. Its Hive-32B model runs end-to-end enterprise workflows such as insurance claims processing and customs duty exceptions without human intervention, and the company claims roughly 10x lower cost per task than frontier APIs. Founded in April 2026 by Ritankar Das with co-founders Reid Hoffman and Mark Pincus, it has signed contracts worth up to $50M in estimated annualized value.

Prox builds AI technical support agents for complex physical products such as power tools, HVAC systems, industrial heaters, EV chargers, and CNC machines. Its multimodal platform is grounded in product manuals and cited to the page, and can draw wiring schematics, share CAD models, and handle voice and video support across product pages, dealer counters, internal tools, ChatGPT, and Claude. The company serves hardware and industrial-equipment makers in powersports, agriculture, HVAC, firearms, and construction equipment.

RaySecur develops terahertz (T-ray) millimeter-wave desktop scanners and AI-powered software for detecting concealed threats and contraband in mail and packages. Its MailSecur imaging systems and 4D-AI threat-detection platform are used by Fortune 500 corporate mailrooms, government agencies, and correctional facilities. The company is expanding into a connected IoT ecosystem for fleet management and threat intelligence.

Rebulk (formerly dScribe AI) uses computer vision and autonomous drones to track bulk inventory such as grain, sand, and gravel in real time. The platform creates 3D reconstructions from drone and fixed camera feeds to deliver precise volume and weight measurements for industrial supply chains, serving customers in agriculture and mining. Part of Y Combinator's S25 batch.

Cloud-based AI-powered image editing platform offering face swapping, video editing, image upscaling, watermark removal, and background removal tools

Sauron is the world's first perceptual home security platform that continuously scans the perimeter of your home and reliably identifies and deters all threats using multi-modal sensor fusion combining high-resolution cameras with LiDAR, radar, and thermal sensors for real-time 360-degree perception.

Segmed provides real-world, diagnostic-grade medical imaging data for healthcare AI and life sciences research. Its platform sources, de-identifies, standardizes and links imaging studies from a network of health systems so innovators can train and validate AI models and support clinical trials. Founded out of Stanford and backed by Y Combinator, the company is based in Palo Alto, California.

Self Inspection builds an AI-powered vehicle inspection and vehicle-condition intelligence platform. Users capture a car through guided smartphone photo capture and the system automatically detects scratches, dents, and tire wear, producing standardized damage estimates and condition reports. It serves dealerships, rental companies, fleets, lenders, auctions, and marketplaces as a continuously updated system of record for vehicle condition, and has completed over one million inspections.

Shofo is building the world's largest indexed video library — described as 'Common Crawl for Videos' — by continuously crawling and labeling public video content from social platforms like TikTok. It sells custom, multi-stage annotated datasets (object detection, activity recognition, semantic segmentation, reasoning annotations) to AI labs for pre-training and fine-tuning. Part of Y Combinator's W2026 batch.

SomaPix is a Rwanda-based photo-sharing platform that uses AI facial recognition to let event attendees instantly find and download photos of themselves from weddings, conferences, sports events and other gatherings. It also gives professional photographers a way to distribute and monetize their event galleries. The company is early-stage, founded by Bonheur Iraguha in Kigali, with a small founding team.

Global generative AI platform automating spatial design for architecture, interior design, and real estate with photorealistic AI renderings, 2D-to-3D automation, and AI virtual staging.

Tera AI develops software-only, platform-agnostic spatial reasoning AI for autonomous robots. Their system enables zero-shot visual navigation across any robot type and environment without hardware modifications, using only existing cameras and onboard GPUs. Founded in 2023 by Tony Zhang (former ML lead at Google X), the company targets robotic manufacturers looking to expand autonomy across diverse platforms.

Throne is building an AI-powered smart toilet device that clips onto standard toilets and uses computer vision and sensors to continuously monitor gut health, hydration, and urological function. The device analyzes stool and urine biomarkers with every flush and delivers daily health insights via a mobile app without requiring manual sample collection. The company has research partnerships with the University of Chicago, MIT, and the Crohn's & Colitis Foundation.

Traini is a Palo Alto-based startup building generative-AI apps and a Cognitive Smart Collar for pets. Its models interpret a dog's emotions and intent from barks, movement, expressions, and physiological signals to enable real-time human-dog communication, health monitoring, and AI pet training. Its emotion-translation accuracy reaches up to 94% across nearly 120 breeds.

Transfyr builds physical AI infrastructure that captures scientific work through sensors and multimodal models. Its platform converts laboratory execution into machine-readable data for reproducibility, automation, training, and technology transfer. The Cambridge, Massachusetts startup operates an in-house wet lab and is hiring technical staff.

Easy-to-use online video editing platform for creating stunning videos, adding subtitles, and growing audience with professional tools

viAct provides AI-powered video analytics and IoT software for construction and heavy industry, helping teams monitor safety, compliance, and operational risk in real time.

Wenti Labs develops AI automation solutions for construction inspection and defect management. The platform allows teams to snap photos on-site and automatically generates inspection reports without manual data entry, streamlining communication and workflows across construction projects.

WolkenVision develops high-precision BIM modeling solutions using AI and 3D LiDAR point cloud technology for construction and infrastructure. The platform offers point cloud management, AI-powered auto-segmentation, and quality inspection software that accurately captures and interprets spatial data for urban development, infrastructure management, and major construction projects.

Yellow is an AI-powered 3D character generation platform that enables professional-quality character modeling through generative AI, backed by Andreessen Horowitz games fund.