Human first. AI where it counts.
We build the foundation as human, rule-based decision engines - transparent, auditable, designed by people who have done feasibility for a living. Then Starlite-AI™, our AI agent, is applied in deliberate phases, only where it genuinely improves time saving and capability. Never as a black box, never overriding your inputs.
How we build
Two layers. One order of operations.
The human foundation comes first - rule-based engines that are correct, explainable, and defensible on their own. It is released and in enterprise use today. Starlite-AI™, our AI agent, is being built on top of that foundation in four phases, with estimated release in 2027. Select any layer to see where it stands and what it does.
Stage 2 | Starlite-AI™
The AI Layer - our AI agent, built in four phases
Early stage | Est. release 2027Purpose-built, rule-based decision engines designed by clinical trial domain experts - EKO™ multi-criteria scoring, Monte Carlo simulation, deterministic timeline derivation. Deterministic and idempotent: the same inputs always produce the same outputs. Transparent, auditable, and correct with no AI in the loop. AI already assists behind the scenes, enriching public registry data and powering natural-language search in Trial Core™.
Held up by five principles
Trusted AI
Purpose-built decision engines designed by clinical trial domain experts - not generic AI bolted onto spreadsheets.
Proven Methodologies
Multi-criteria decision analysis, Monte Carlo simulation, deterministic scheduling - methods trusted for decades in finance, aerospace, and operations research.
Tested Technology
Built alongside active clinical teams and proven on live projects - RFPs, bid defences, and study planning with global CROs.
Auditability
Every recommendation traces to its inputs, weightings, and evidence - with version history and audit trail that hold up in front of sponsors, regulators, and boards.
Transparency
The engines never override your inputs or make autonomous decisions. Before any planned AI feature ships, its privacy assessment is re-run and the public documentation updated.
AI accelerates the workflow. The rules, the methodology, and the judgement stay visible - and stay yours.
The development approach
Built with the industry, not in isolation.
Aurora is primarily people from clinical research, and we leverage that internal experience in every design decision. Our tech team trains on the key concepts and workflows of clinical trials, so engineering and domain expertise pull in the same direction - the synergy that designs better products. Partners, collaborators, and customers then refine them further on live projects, and are key to finding even more value.
400+
Studies of feasibility experience across the team, behind every engine design
Clinical-first
A team primarily from clinical research, with engineers trained in trial concepts and workflows
Live projects
Refined with partners, collaborators, and customers on real RFPs, bid defences, and study planning
ESMO 2024
Methodology peer-reviewed at a leading oncology congress
The evidence
Read the proof behind the product.
See the engines - and the evidence - for yourself.
Book a demo and we'll walk through the methodology behind every output, on an opportunity you already know.