EvidLabs is building what is effectively a domain-specific agentic workflow for scientific synthesis. Systematic reviews are inherently agentic: they require a set of instructions (inclusion criteria), a search phase, a filtering phase, and a data extraction phase. EvidLabs automates these steps, moving toward a world where a "Researcher Agent" can autonomously generate a draft systematic review for human verification.
In the broader agent ecosystem, EvidLabs represents the verticalization of agents in high-stakes environments like medicine. They are active in the "knowledge worker" layer of the stack, specifically focusing on the transition from unstructured text (scientific papers) to structured evidence. Their work is a prime example of how LLMs are being tuned for high-precision, citation-heavy tasks where the cost of error is high.
The volume of scientific literature is growing at a rate that exceeds the human capacity to synthesize it. In fields like medicine and life sciences, the "gold standard" for truth is the systematic review—a rigorous process of identifying, appraising, and synthesizing all relevant research on a specific topic. Historically, this is a manual, multi-month endeavor involving thousands of papers, multiple human reviewers, and significant risk of oversight. EvidLabs focuses on this specific bottleneck, applying machine learning to the specialized domain of evidence-based research.
EvidLabs operates through its primary platform, Revidium. The company identifies a core problem in the current research cycle: the transition from raw search results to actionable evidence. While tools like PubMed or Google Scholar are effective at finding papers, they do nothing to help a researcher actually process them. Revidium is designed to sit between the search phase and the final synthesis, providing a cloud-based environment where AI handles the repetitive tasks of screening and data extraction.
A typical systematic review requires a researcher to screen thousands of titles and abstracts against strict inclusion criteria. EvidLabs uses language models to assist in this screening process, identifying relevant papers with a higher degree of speed than manual review. Beyond simple keyword matching, the platform is built to understand the context of research questions—often formulated using the PICO (Population, Intervention, Comparison, Outcome) framework.
After screening, the platform moves into data extraction. This is where researchers pull specific data points—sample sizes, dosage, outcomes, and p-values—out of PDFs and into a structured table. This is traditionally the most error-prone part of a review. By automating this extraction, EvidLabs aims to reduce the time-to-publication for researchers while maintaining the audit trail required for academic publishing. The company emphasizes that its tools are for "evidence-based" work, a distinction that implies a higher threshold for accuracy and citation than general-purpose AI summarizers.
EvidLabs enters a market currently dominated by established players like Covidence and Rayyan. These legacy tools provide the project management structure for reviews but often lack the deep AI integration required to actually perform the reading tasks. EvidLabs is part of a newer cohort of AI-first research tools that treat the paper as a structured data source rather than just a document.
The company is registered as EvidLabs (Private) Limited, indicating a focused corporate structure likely geared toward the healthcare and pharmaceutical sectors, where the demand for rapid evidence synthesis is highest. By focusing on the "meta-analysis" and "systematic review" keywords, they are targeting a high-intent user base: academic researchers and medical librarians who are legally or professionally required to follow specific methodological standards. The challenge for EvidLabs lies in proving that AI can meet the rigorous "Risk of Bias" assessments required in medical research without introducing hallucinations or missing critical negative results.
A platform for conducting systematic reviews and meta-analysis.
EvidLabs is hiring