DeepMirror Bridges AI's 'Reasoning-to-Action' Gap

AI firm DeepMirror has integrated the OpenClaw framework into its Physical AI stack. The move aims to close the gap between an AI's ability to reason and its ability to execute physical tasks, a key challenge in robotics. This follows a profile on OpenClaw's founder, Peter Steinberger, and his journey from solo indie dev to open-source leader.

DeepMirror, a London-based AI drug discovery startup, recently secured $2.4M in pre-seed funding led by Twinpath Ventures, supplemented by an Innovate UK grant. The company's platform aims to accelerate the design of new drug molecules, a process that can take up to six years and cost over a billion dollars. DeepMirror's focus is on making AI for molecule design accessible to smaller pharma companies without requiring them to build large in-house AI teams or give up intellectual property. The OpenClaw framework, created by Austrian software engineer Peter Steinberger, is an open-source platform for developing and running personal AI agents. It allows an AI to interact with various software applications to automate tasks. Steinberger, who previously founded and bootstrapped the successful software development kit company PSPDFKit, started OpenClaw as a personal project in April 2024 before its public release around November 2025. Following the viral popularity of OpenClaw, Steinberger joined OpenAI in February 2026 to lead the development of personal AI agents. However, the OpenClaw project itself was not acquired and will transition to an independent foundation to ensure it remains open-source. Steinberger's goal is to accelerate the creation of an AI agent that is usable by a non-technical audience. The integration of OpenClaw enables DeepMirror's "Physical AI" to give robots a "world memory," allowing them to understand and remember objects, people, and events in relation to space and time. This capability, termed Spatial Agent Memory, moves beyond scripted actions, allowing robots to perform tasks with greater autonomy, such as optimizing inventory in logistics or carrying out maintenance in hazardous environments. This addresses a key challenge in robotics: bridging the gap between an AI's high-level reasoning and its ability to execute tasks in the complex, ever-changing physical world. The NYC AI scene is rapidly expanding, with major tech companies like Google DeepMind, Meta, and OpenAI establishing a significant presence. The city is also home to a growing number of AI startups, including Runway, Hebbia, and Assembled, which are hiring for various roles. Venture capital firms like Lux Capital are actively investing in the local ecosystem, supporting companies such as Hugging Face and Modal. VC funding for AI startups has seen a dramatic increase, accounting for over half of all venture capital investments in 2025. In 2024, AI startups raised a third of all venture capital, with seed-stage AI companies commanding a 42% higher valuation than their non-AI counterparts. This trend is driven by major funding rounds for companies like OpenAI and xAI, and significant investments from firms like Andreessen Horowitz and Sequoia Capital. The rise of Vertical AI SaaS, which creates AI-powered solutions for specific industries, is transforming various sectors. By embedding AI into specialized workflows, these platforms can automate complex tasks, leading to significant cost reductions and efficiency gains. This allows for new business models that charge for outcomes rather than user seats, turning software into an engine for specific results.

Get your own daily briefing

Scout delivers personalized news, insights, and conversations tailored to your role and industry.

Download on the App Store

Shared from Scout - Be the smartest in the room.