Skip to content
2026ContactGet updates
All ten talk summaries

SDLCAI 2026 / Talk summary

Jenni Kylmäkoski

Why Leaders Should Use AI Themselves and Eat Their Own Dog Food?

Jenni Kylmäkoski explains why leaders should build and use AI tools themselves. Her reporting tools and an unintended Jira change reveal practical lessons about onboarding, permissions, observability, learning time and maintaining shared AI assets as products.

25:30 · 13 October 2026

Watch full recordingProgramme & current materials
  • leadership
  • AI adoption
  • developer experience
  • least privilege
  • observability
  • learning
  • AI asset maintenance

Takeaway graphic

Share three selected takeaways from this talk. Download a landscape or portrait card, with suggested post copy and image alt text.

Leaders need hands-on experience. Jenni Kylmäkoski at SDLCAI 2026. Three paraphrased talk takeaways: 1. Build tools for your own work and validate the outputs. 2. Give permissions, logs and recovery explicit attention. 3. Maintain shared AI assets as software products. sdlcai.org
Three paraphrased takeaways, with timestamps from the final recording.
Landscape PNG · 1200 x 630Portrait PNG · 1080 x 1350Download post copy & alt text
Read post copy & alt text

Suggested post copy

Leaders need hands-on experience Three takeaways from Jenni Kylmäkoski at SDLCAI 2026: 1. Build tools for your own work and validate the outputs. 2. Give permissions, logs and recovery explicit attention. 3. Maintain shared AI assets as software products. Paraphrased from “Why Leaders Should Use AI Themselves and Eat Their Own Dog Food?”. Watch the full talk: https://www.youtube.com/watch?v=1bT23mhlzGg #SDLCAI

Image alt text

Leaders need hands-on experience. Jenni Kylmäkoski at SDLCAI 2026. Three paraphrased talk takeaways: 1. Build tools for your own work and validate the outputs. 2. Give permissions, logs and recovery explicit attention. 3. Maintain shared AI assets as software products. sdlcai.org

Key points

Paraphrased from the talk. Follow each timestamp to hear the argument in context.

  1. Leaders can learn the development experience by building tools for their own recurring work and validating the outputs.

    5:41–9:07

  2. Onboarding, documentation, training and support shape whether shared AI assets are actually used.

    9:07–11:34

  3. A model change can alter tool behaviour; least privilege, logging and recovery deserve explicit attention.

    16:00–19:09

  4. Learning time must be included in priorities and delivery plans, or urgent commitments will displace it.

    19:20–20:50

  5. Shared AI assets need a product lifecycle with requirements, tests, releases, documentation and maintenance.

    20:50–22:57

The talk in context

Drawing on her work as an R&D leader at Nokia, Jenni Kylmäkoski argues that leaders need direct experience of the changes they ask teams to make. She builds small applications for recurring reporting and analysis, using the same tools as her developers and checking the generated results. This exposes installation friction, documentation gaps and the support people need to adopt new capabilities. A model change unexpectedly reopened closed Jira tickets during a reporting task, prompting her to reconsider permissions, logs and recovery. She also describes the tension between delivery commitments and time spent learning. Her conclusion is that an organisation's prompts, skills, agents, workflows and monitoring need requirements, testing, distribution, documentation and ongoing maintenance, just as other software products do. Practical participation helps leaders understand both the technology and the people living through the transition.

Continue with another talk

Explore related questions across the seminar.