Work in progress— changes will come
Ignite26/27
Back to all phases
Phase 04

January–February 2027

Launch

Ship is about staying alert while it runs. You learn how AI solutions behave in production — where they fail, how you see it before the client does, and how to roll back without panic. At the same time, you learn to balance cost, performance and quality under real constraints.

Learning objectives

  • Operate AI solutions reliably in production
  • Establish observability that reveals failures and cost
  • Roll out changes safely with canary + rollback
  • Secure the solution with threat modelling and controls

Core topics

  • 01GenAIOps: observability, monitoring, alerting
  • 02Canary deploy, feature flags, rollback strategies
  • 03Model router and fallback logic
  • 04Cost control: caching, batching, semantic cache
  • 05Security: STRIDE, prompt injection defence, secret management

Technologies and frameworks

  • Langfuse dashboards · OpenTelemetry · Grafana
  • LaunchDarkly · GrowthBook (feature flags)
  • Vercel · Railway · Docker
  • STRIDE modelling · OWASP LLM Top 10

Sessions in this phase

  • Masterclass: 'where does it burn most?' — observability
  • Lab weekend (Q3): 'build an SLO dashboard'
  • Dojo on prod incidents from the cohort
  • Frontier Friday

Artefacts Ignitere deliver

  • 01Observability dashboard with SLOs and alerts
  • 02Runbook for incident response
  • 03Cost model for AI delivery with documented trade-offs
  • 04Threat model for an agentic solution

Courses in this phase

Shared coursesAll Ignitere

  • GenAIOps og observability

    Gi innsikt i hva som skjer i produksjon — og fang feilene før kunden.

  • Kost-kontroll og model routing

    Send riktig forespørsel til riktig modell. Reduser kost uten å miste kvalitet.

  • Sikkerhet: prompt injection og STRIDE

    Trusselmodeller en agentløsning som om en motstander prøver å ødelegge den.

Track courses · Ignite Builder

  • Semantic cache og prompt-cache-strategier

    Spar 30–80 % av token-kost uten å miste kvalitet.

  • Container og cloud-deploy for agenter

    Fra lokalt prosjekt til skybasert produksjon.

Track courses · Ignite Tuner

  • Fine-tuning og distillation

    Når det lønner seg å trene, og når du bare skal la være.

  • Natural-language-to-SQL og agentic analytics

    La modellen stille spørsmål direkte til dataen — trygt.

Cross-cutting threads

  • AI economics and FinOps
  • EU AI Act practice
  • Consultant craft
Phase 03Craft
Phase 05Anchor