High-Level Overview
Automat is an AI-powered robotic process automation (RPA) platform that enables businesses to automate repetitive tasks using natural language instructions, screen recordings, or videos, making it 10X faster and 1/10th the price of traditional tools like UiPath.[2][3][4] Founded in 2023 by ex-Googlers Gautam Bose and Lucas Ochoa, it targets operations teams in sectors like insurance and fintech, solving the problem of slow, expensive enterprise automation by leveraging large language models (LLMs), computer vision, and self-healing agents that handle exceptions with human oversight.[2][3][4] The company has gained strong growth momentum, raising a $15.5M Series A in November 2025 led by Felicis (with Initialized, Khosla, Y Combinator, and others), following earlier funding, and serving customers like AmeriTrust and Ziina with custom automations for document processing, data extraction, and workflow integration.[2][4]
(Note: Search results reference an unrelated earlier company named Automat, founded in 2016 and acquired in 2022 by Salesfloor for conversational AI on messaging platforms; this profile focuses on the active 2023 YC-backed Automat at runautomat.com.)[1][3]
Origin Story
Automat was founded in 2023 by Gautam Bose (CTO) and Lucas Ochoa, who met on the first day of design school at Carnegie Mellon University and have been roommates since.[2][3] Their idea emerged from hands-on experience at Google in early 2020, where they explored product applications for LaMDA, building over 50 prototypes and contributing to its first public demo (AI Test Kitchen).[2] Spotting that enterprise automation relied on outdated, costly tools with high licensing fees (e.g., $66K/year for UiPath) and consultant setups ($50K+), they applied AI techniques from robotics to software automations.[2][3] Early traction came via Y Combinator, where they offered waived setup fees to the first 10 YC companies and built automations from customer videos, such as scraping emails, prospecting, and spreadsheet saves.[3] Based in San Francisco's Mission District, the team of robotics/AI engineers quickly iterated from prototypes to a managed service platform.[2][3]
Core Differentiators
- AI-Driven Natural Language Automation: Users describe tasks or share screen recordings/videos; Automat's proprietary transformer-based RPA engine uses LLMs and computer vision to build self-healing agents that mimic human behavior on Chrome, Windows, Citrix, or web portals—without coding or APIs.[3][4]
- Cost and Speed Advantage: 10X faster and 1/10th the price of legacy RPA, with no massive teams or budgets needed; includes triggers via API/UI and a Sandbox for non-technical users.[2][3][4]
- Reliability with Human-in-the-Loop: Agents reason through exceptions, learn from feedback for precision, and offer built-in oversight, ensuring security and consistency for complex tasks like document processing or filings.[2][4]
- Developer and Operator Experience: Forward-deployed engineering acts as an extension of customer teams, with fine-tuning (e.g., GPT-4o for industry-specific IDP/RPA) and proven results in real-time verification/activation.[2][4]
- Backed Ecosystem: Y Combinator alumni with top-tier investors (Felicis Series A lead), fostering rapid scaling and community trust.[2][3]
Role in the Broader Tech Landscape
Automat rides the agentic AI wave, where LLMs and computer vision enable "AI agents" to handle knowledge work autonomously, shifting from rigid RPA to adaptive, natural-language-driven automation amid exploding demand for efficient enterprise ops.[2][3] Timing is ideal post-2023 LLM breakthroughs (e.g., ChatGPT, LaMDA insights), as market forces like labor shortages, rising costs, and no-code trends favor cheaper alternatives to $50K+ consultant projects.[2][3] It influences the ecosystem by democratizing RPA for non-technical ops teams in insurance/fintech, unlocking revenue via streamlined processes (e.g., policy data extraction sans APIs), and pushing boundaries with vision fine-tuning—positioning it against incumbents while amplifying YC's AI automation portfolio.[2][3][4]
Quick Take & Future Outlook
Automat is primed to dominate AI-native RPA, with its $15.5M Series A fueling Sandbox expansions, broader platform triggers, and vertical-specific fine-tunes (e.g., LOS integrations, GPT-4o enhancements).[2] Trends like multimodal AI agents and "agency for people" (human-authored bots) will shape its path, potentially capturing share from legacy players as enterprises prioritize 10X efficiency gains.[2][3] Its influence may evolve into a full enterprise platform, empowering ops teams globally and redefining automation as accessible as screen-sharing a task—tying back to its core promise of turning manual drudgery into seamless AI agency.[2][4]
