Aviro
About
Aviro builds long-horizon reinforcement-learning environments, post-training datasets, and benchmarks for coding, computer use, knowledge work, and research. Its environments expose agents to complex multi-step workflows designed for model training and evaluation.
What They Offer
Commercial or operational services this company provides.
Agent evaluations
Custom RL environments
Training datasets
Products & Public Artifacts
Ebla-1
Benchmark · Public
Grounded-reasoning benchmark for evaluating advanced model and agent capabilities.
Visit artifact →Long-horizon RL environments
Environment · Commercial
Simulated environments spanning coding, computer use, knowledge work, and research, designed to train agents through long-horizon reinforcement learning and post-training.
Visit artifact →Technical Capabilities
Code / Terminal ExecutionBrowser UseComputer UseLong-Horizon TasksReal-App / Workplace SimulationTool / API / MCP Use
Only publicly documented capabilities are shown. Absence does not imply the capability is unavailable.
Focus Areas
Computer UseCodingEnterprise AgentsRL EnvironmentsTraining DataRLHF / Post-trainingEvaluations
Domains
Software EngineeringComputer UseScienceEnterprise
Leadership
Ayush Karupakula
Founder / COO
Aarush Kukreja
Founder