Ignite26/27
The Program

Six phases from diagnosis to delivery

The structure is fixed, while the content is updated continuously to match the market. Each phase has its own learning objectives, sessions and artefacts, building step by step towards what consultants deliver for clients.

  1. Phase 01September 2026

    Foundations

    Onboarding 1–4 September, responsible AI and shared LLM foundations.

    Foundations establishes the cohort. Onboarding on 1–4 September lays the groundwork: the cohort meets, teams form and expectations are set. The month holds the two shared courses the rest of the year rests on — what a language model actually does, and what responsibility comes with putting one into production at a client.

    Læringsmål

    • Establish shared language and mental models for working with language models
    • Explain where an LLM fails — hallucination, bias and drift
    • Know the responsibility that follows an AI delivery: traceability, risk classes and client confidentiality
    • Build cohort dynamics and working habits that last the year

    Kjerneinnhold

    • Responsible AI: risk, traceability and the EU AI Act in practice
    • LLM Foundations: how the models work, and prompt patterns
    • Onboarding: team formation, expectations and shared standards
    • Dojo and Frontier Friday get under way
    Se samlinger og artefakter for Foundations
  2. Phase 02October 2026

    Discover

    The toolkit, the tracks begin, and the economics behind an AI initiative. Weekend gathering 1, 17–18 October.

    Discover is about finding the right problem — and about getting hold of the tools to attack it. The cohort learns AI-assisted development, sets a shared standard for what "ready" and "done" mean, and works out what an AI initiative actually costs. At the weekend gathering on 17–18 October the cohort splits into tracks for the first time.

    Læringsmål

    • Use AI-assisted development in your own work — and know when it does not pay off
    • Define DoR and DoD for your own team
    • Work out what an AI initiative costs, and choose a run pattern that holds the budget
    • Present an AI proposal to a client in person

    Kjerneinnhold

    • AI-assisted development: Claude Code, Cursor and the agentic IDE
    • Team standard: DoR and DoD for AI delivery
    • AI economics and FinOps: token cost, caching, routing
    • Presentation skills and client communication
    • Track breakouts: multi-provider (Builder), model selection (Tuner), AI readiness (Shaper)
    Se samlinger og artefakter for Discover
  3. Phase 03November–December 2026

    Build

    RAG, evaluation as its own discipline, and the EU AI Act as the frame around what gets built.

    Build is the technical core. You implement RAG against real data and establish an eval pipeline that catches regressions before they reach the client. At the same time you learn to classify the solution under the EU AI Act and to justify it in a business case — technical quality that cannot be defended commercially rarely gets bought.

    Læringsmål

    • Implement RAG against enterprise data with measurable quality
    • Establish an eval pipeline that runs in CI
    • Classify a solution under the EU AI Act and document the assessment
    • Build a business case for an AI initiative

    Kjerneinnhold

    • RAG engineering: chunking, retrieval, re-ranking, hybrid
    • Eval as a discipline: golden sets, rubrics and LLM-as-judge
    • Eval in CI — regressions caught before delivery
    • EU AI Act: risk classification, DPIA and documentation
    • Business case modelling for AI delivery
    • Robust AI services: retries, circuit breakers, idempotency (Builder)
    Se samlinger og artefakter for Build
  4. Phase 04January–February 2027

    Operate

    Agents are assembled, connected to enterprise systems and put into production — with observability, cost control and security.

    Operate is about staying alert while it runs. This is where the agents are assembled, connected to enterprise data via MCP and put into production. You learn where they fail, how you see it before the client does, and how to roll back without panic — while balancing cost, performance and quality under real constraints.

    Læringsmål

    • Build an agent that plans, uses tools and handles state
    • Connect agents to enterprise systems via MCP with sound access control
    • Establish observability that gives insight into failures, latency and cost
    • Roll out changes safely with canary, feature flags and rollback

    Kjerneinnhold

    • Agent orchestration: planning, tool use, state management
    • MCP integrations against enterprise data and actions
    • GenAIOps: traces, observability, alerting
    • Rollout: canary, feature flags, rollback and kill switch
    • FinOps for AI: model routing, caching and distillation
    • Security and agent governance: injection defence, STRIDE, agent IAM
    Se samlinger og artefakter for Operate
  5. Phase 05March–April 2027

    Adopt

    Change management, user research, adoption metrics and handover — and the portfolio starts to take shape.

    Adopt is about turning a working solution into one that actually gets used. You learn user research, change management and how to measure real usage, and how a good handover lets the client keep running it. Weekend gathering 2, 13–14 March, runs The Phoenix Project as a DevOps simulation. Towards the end of the phase, work on your own portfolio begins.

    Læringsmål

    • Design an adoption strategy for an AI initiative
    • Run user research and measure actual usage
    • Facilitate workshops that anchor change
    • Design a handover that secures the client's continued operation

    Kjerneinnhold

    • Change management: Prosci ADKAR
    • User research and adoption metrics: WAU, engagement depth, feedback loops
    • Workshop facilitation and executive communication
    • Handover, operational transfer and AgentOps
    • Data governance for AI: lineage and consent (Tuner)
    • Portfolio: selection, narrative and visibility
    Se samlinger og artefakter for Adopt
  6. Phase 06May–June 2027

    Graduate

    Alliance case, oral defence and graduation. In-person gathering 16–17 June.

    Graduate is where it all comes together. You assemble ten months of learning into one coherent delivery: the Alliance case. Then you defend it orally before a jury and graduate as an Ignite Certified Consultant at the gathering on 16–17 June.

    Læringsmål

    • Deliver the Alliance case with the program's standards built in
    • Defend the delivery orally before a jury
    • Present the portfolio as decision material in sales
    • The transition: from Igniter to Ignite Certified Consultant

    Kjerneinnhold

    • Final delivery: a complete AI system with evals, observability, cost model and documentation
    • Oral defence: pitch, demo and handling questions
    • Norwegian sector advisory (Shaper)
    • The road ahead: how the title is used in sales and delivery
    Se samlinger og artefakter for Graduate

Explore the program further

Two ways to dig deeper into what the year contains.

Course catalogue

The full course catalog — the shared courses everyone takes, plus dedicated track courses for Builder, Tuner and Shaper. Click around and filter by track.

Course catalogue

Sessions

The rhythm that carries the year. Masterclass, Dojo, Frontier Friday, lab weekends and the two physical gatherings.

Sessions
Juni 2027

Graduation

To dagers avslutning: pitch av leveranser til kundepanel, sertifiseringsseremoni, og Alliance-partnere inviteres inn. Her markerer vi overgangen fra Igniter til Certified Consultant.

Se detaljer for graduation
Next step

See which track fits the client

Each phase splits between shared and track courses. The tracks explain who takes what.

See the tracks