README.md

iml4e

Reaktor's training material contributions on ITEA IML4E project.

MLOps team excercises

The following excercises are suitable for a solution team of developers and ML specialists. It is advisable to go first thorugh an introdcution to MLOps technical framework before the excercises. For example Helsinki University course uses this material as excerices. The excercises are meant to be done as team discussions and planning.

• Excercises slides and summary of expected themes in team answers (pdf and pptx). See ./docs/

Hints

• Usually solutions have upsides and downsides for the various stakeholders. Analyse them. Assess if the solution would be feasible in the real world. If not, what would you deem the hardest obstacle? Stakeholder analysis: who needs what, who has to do things? Even some simplified business canvas work could help? Or a DevOps adaptation?
• Should the solution be essentially technological or cultural?
• Is a general plan enough? Should you specify the particular tools? Should you implement an example?

Material for designers: crafting an LLM system

• Crafting LLM powered interactions: design principles for natural language user interfaces
• Designing disruptive GenAI services

More whitepapers and articles

On MLOPs, building AI/ML/GenAI systems

• Why some ML/AI projects succeed while others fail
• Data meets business - lessons learned the hard way
• Mastering a GenAI transition from PoC to production
• Things we have learned developing GenAI applications
• LLM case in healthcare
• MLOps making machine learning production work

On design

• AI is a solution in search of-a-problem
• Train yourself not the model: a designer's view on language models