Reported impact
Business outcome
OnticWorks.io used contextual AI to reclaim manager time, cut agricultural water use and manual monitoring, improve crop yields, accelerate its robotics education launch, and help students deploy autonomous agents from simulation to physical hardware more successfully.
- Cost Reduction
- Productivity Improvement
- Speed to Market
Solution
OnticWorks.io integrated IBM Bob into Elmer's core reasoning loop so the platform could attach spatial context to the multimodal information it was already processing. Cameras, environmental sensors, drone systems, enterprise applications, and user interactions feed the system, while IBM Bob models the relationships between people, places, assets, and events as they unfold. IBM watsonx.ai supplies the multimodal reasoning layer and NVIDIA-powered inference runs on Red Hat OpenShift, allowing Elmer to combine voice, vision, gesture recognition, and operational telemetry before producing a recommendation. Recommendations are then routed through a draft-and-approve workflow so humans remain in control and the system learns from overrides and approvals. OnticWorks.io reused the same architecture across administrative operations, agricultural decision support, and robotics education rather than building separate products for each environment.
Data flow
Webcams, sensors, drones, enterprise applications, and user interactions feed Elmer; IBM Bob enriches those inputs with spatial relationships, IBM watsonx.ai reasons over the multimodal context on Red Hat OpenShift, and the platform returns draft recommendations that users approve, modify, or override.