Daniel Darling is a former advertising technology founder turned venture capital investor at Focal who hosts the "Five Year Frontier" podcast, where he interviews founders from companies like Clari, Voltron Data, and Poolside to explore how their products will reshape industries over a five-year horizon. As an Australian-born, San Francisco-based investor, he serves as a bridge for international capital (especially Australian investors) seeking access to early-stage AI and software startups in North America, while providing company-building support beyond capital in areas like go-to-market strategy and product positioning.

• A focus on the "Agentic Enterprise" and the shift toward autonomous, always-on workforces. He frequently highlights how AI agents will move beyond simple tasks to running complex functions 24/7, noting that agents can deliver over four times the annual output of human employees and citing companies like Ema that are building the infrastructure for this transition.
• Deep interest in the evolution of software development from AI-assisted to AI-led. He posits that as code generation accelerates via tools like Cursor, the critical bottleneck shifts to quality control and code review, specifically praising Graphite for solving this choke point for the next generation of high-velocity developers.
• Enthusiasm for "hard tech" that bridges the digital and physical worlds, particularly in energy, biotech, and materials science. This includes transformative ideas like converting carbon dioxide into rocket fuel for interplanetary travel (referencing Air Company) or using AI to simulate advanced materials (citing Orb Materials) to redesign physical infrastructure.
• Strong conviction in open-source AI and the "small model revolution" overtaking proprietary closed models. He argues that the future belongs to collective innovation and efficient enterprise deployment rather than just massive closed labs, often discussing how open innovation is rewriting the playbook for unit economics.
• Investment in simulation and synthetic data as the key to unlocking autonomy in robotics and vehicles. He argues that the race for self-driving isn't about collecting billions of real-world miles but using unsupervised learning to train systems smarter and faster, pointing to Helm AI’s approach to leapfrogging traditional fleet-based data collection.
• A vision for the future of compute infrastructure that moves away from brute-force GPU stacking. He speculates on radical changes to data center design, predicting they may eventually resemble biological systems, utilize liquid cooling, or even be powered by small nuclear reactors to handle the immense energy demands of AI.
• Advocacy for voice as the primary user interface for the AI era. He suggests that controlling voice input allows for better command and structure of data, viewing it as an interactive command center rather than just a way to dictate text, effectively acting as a wedge to introduce AI agents into business workflows.