New AI Lab Core Automation 'Nerdsniped' Researchers From Anthropic, Google DeepMind

Why It Matters

The talent shift gives Core Automation a competitive edge in pioneering automated AI research, potentially accelerating breakthroughs beyond the scaling‑focused approaches of big tech. It also intensifies the talent war, prompting larger firms to rethink retention and compensation strategies.

Key Takeaways

• Core Automation founded by ex‑OpenAI VP Jerry Tworek
• Team includes former Anthropic and DeepMind researchers
• Goal: automate AI research with self‑building systems
• Reflects rising trend of talent moving from big labs to startups
• Startup offers equity, co‑founder titles to attract top talent

Pulse Analysis

The AI industry is witnessing a pronounced talent migration as leading researchers abandon heavyweight labs for nimble startups. Recent years have seen giants like Meta, Google, and OpenAI invest billions to retain staff, yet the allure of ownership, faster decision‑making, and deeper scientific impact draws talent toward smaller ventures. Core Automation’s formation epitomizes this shift, gathering a cadre of engineers who helped shape frontier models at Anthropic and DeepMind. Their departure signals that even well‑funded labs may struggle to keep the most innovative minds when the promise of pioneering new research paradigms is on the table.

Core Automation’s mission to "automate the process of building" AI systems marks a strategic departure from the prevailing model‑scaling paradigm. By focusing on novel learning algorithms and architectures that can self‑optimize, the startup aims to reduce the manual engineering bottlenecks that dominate current research cycles. If successful, such automation could dramatically shorten development timelines, lower compute costs, and democratize access to cutting‑edge models. The company’s early messaging suggests a vision where research teams spend less time on incremental scaling and more on conceptual breakthroughs, potentially reshaping how AI labs operate.

The broader market implications are significant. As startups like Core Automation lure elite talent with equity stakes, co‑founder titles, and dedicated compute resources, big tech may need to augment salary packages with more substantive research freedom and intellectual property incentives. Moreover, investors are likely to view automated‑research platforms as high‑growth opportunities, spurring additional capital into the AI startup ecosystem. This talent‑driven competition could accelerate the diversification of AI approaches, fostering a richer landscape of models beyond the current dominant large‑scale architectures.