Here are 10 early-to-mid-stage companies building or enabling autonomous scientific experimentation (self-driving labs), avoiding the most famous legacy players (like Emerald Cloud Lab or Strateos) in favor of niche, emerging, or highly specialized startups.
1. Chemify — Digitizes chemistry into executable code ("chemputation") for automated, robotic chemical synthesis. Why it fits: Series A ($43M led by Triatomic). Strong founder signal (Lee Cronin) and a unique approach that treats chemical synthesis like compiling software, creating a true closed-loop autonomous chemistry lab.
2. LabGenius — Uses a robotic platform driven by active learning algorithms to autonomously design, synthesize, and test novel proteins. Why it fits: Series B. Highly regarded in the UK bio-ecosystem with strong traction in pharma partnerships; they have successfully built a pure closed-loop system where AI dictates the next physical experiment without human intervention.
3. Atinary Technologies — Provides a no-code AI software platform (SDLabs) that integrates with existing lab instruments to turn traditional labs into self-driving ones. Why it fits: Seed/Early stage. Strong signal because they are selling the "picks and shovels" (the AI/hardware integration layer) rather than building a proprietary lab, allowing legacy R&D teams to automate their workflows.
4. BioLoomics — Combines automated microfluidics and machine learning for the autonomous directed evolution of antibodies and protein degraders. Why it fits: Seed stage (backed by Innovation Endeavors). Perfectly fits the thesis by combining closed-loop AI with physical high-throughput screening to solve the specific bottleneck of degrading complex targets.
5. Mytos — Builds fully automated, robotic hardware for mammalian cell culture, handling everything from feeding and passaging to imaging. Why it fits: YC alumni, Series A. Strong traction addressing one of the most labor-intensive, error-prone bottlenecks in biotech, effectively automating the foundational step of biological experimentation.
6. DeepCure — Develops AI for small molecule discovery backed by a custom-built, fully automated robotic chemistry lab to synthesize and test compounds. Why it fits: Series A/B. Strong signal in bringing physical synthesis entirely in-house; they realized AI is bottlenecked by physical data generation, so they built a robotic lab specifically to feed their generative models.
7. Tierra Biosciences — Offers an automated, cell-free protein synthesis platform accessible via a digital API. Why it fits: Series A. Strong traction with thousands of proteins synthesized; they act as an autonomous cloud lab specifically for custom protein generation, bypassing the complexities of living cell cultures.
8. Unnatural Products — Pairs machine learning with an automated chemistry platform to design and synthesize cell-permeable macrocyclic peptides. Why it fits: Series A (backed by ARTIS Ventures). Excellent niche focus; macrocycles are notoriously difficult to synthesize, and their autonomous platform creates a massive proprietary data moat in an underserved drug class.
9. Iktos — An AI drug discovery company that recently launched an end-to-end automated robotic lab (Iktos Robotics) for de novo chemical synthesis. Why it fits: While the core company is mid-stage, their new robotics division represents a pure "self-driving lab" pivot, showing strong commercial traction in closing the Design-Make-Test-Analyze (DMTA) loop for external pharma clients.
10. Culture Biosciences — Provides cloud-connected, automated bioreactors that allow synthetic biology companies to run and monitor fermentation experiments remotely. Why it fits: Series B. Proven traction acting as the "AWS for biomanufacturing," enabling companies to run high-throughput, autonomous scale-up experiments without building physical pilot plants.
***
2 Most Interesting White-Space Gaps
1. Hardware-Agnostic Middleware ("Android for Lab Robotics") Every self-driving lab startup currently spends massive amounts of capital building custom integrations between their AI agents and proprietary lab hardware (liquid handlers, mass spectrometers, etc.). There is a glaring gap for a universal, open-API middleware layer that standardizes commands across different hardware OEMs, allowing AI scientists to "plug and play" with any lab equipment.
2. Autonomous Materials Science (Beyond Bio/Pharma) 90% of the self-driving lab ecosystem is focused on drug discovery (proteins and small molecules). There is a massive, underfunded white space for autonomous labs focused on hard-tech materials science—specifically the closed-loop discovery of next-generation battery electrolytes, sustainable polymers, and semiconductor materials.