Kortical Reviews

1) Easy to use interface for managers, business analysts, and data scientists.

2) Interoperable through API, for example easy integration to your Qlik Sense and Power BI dashboards. Where you can visualise all the results and integrate them in your day-to-day decision making.

3) Scalable to any company size through the power of Google Cloud. From small to large-scale enterprises.

4) A data science tool built by senior data scientists and can be trusted.

5) Not a black-box. Each row/result is explainable which is essential to most stakeholders.

1) Would love to see data model governance dashboard. Performance over time. Data drifting.

2) A detailed documentation and guideline to help users.

1) Focus on the business goals and have a lot of meetings to maximize ROI.

2) Be open to change through AI and machine learning.

1) Optimize nurse staffing, scheduling, and planning.

2) Optimize stock inventory for pharmaceutical and medical consumables.

3) Reduce readmission rates.

Kortical's auto-ML platform accelerates the time it takes to move from explorational to deployment as a data scientist. By automating the vast majority of steps in a machine learning pipeline, Kortical enables a data analyst to build production ready machine learn solutions. There are a wide array of supervised learning algorithms to tackle different problems.

For the more advanced data scientists, there is probably a desire to hand craft machine learning solutions from scratch and to be incomplete control of the workflow. There is also a need to conduct most feature engineering prior to ingesting data into the platform.

If you are looking for a machine learning platform for analysts who don't have PhDs in AI or machine learning, Kortical is an excellent platform for demoncratising the availablity of AI/ML solutions. It builds models and makes deployment easy. You will probably still do you data preparation in advance of using the platform.

The most problem we are solving is the ability to build prototypes with little domain knowledge on a particular subject. This unlocks the benefit of reducing time to production (reducing effort 3-10x). It permits more time to experiment at the beginning of a project and there is no time wasted in provisioning the appropriate cloud environment. Typical business problems include binary classifiers for event prediction (cross-sell, churn) and regressors for forecasting.