Five phases. One methodology.
Every engagement — a $3k readiness kit or a $350k transformation — runs the same disciplined arc. Start at any phase.
Diagnose → Train → Innovate → Build → Sustain
Diagnose
Before anything is built or bought, we measure where you actually stand — data, people, infrastructure, and governance.
Full playbook ›Hide playbook
Establish a credible, evidence-based starting point for AI transformation — so every subsequent investment has a defensible reason.
Inputs we work with
- ·Executive interviews across 4–8 stakeholders
- ·Data inventory and data-quality scan
- ·Process maps for in-scope workflows
- ·Current tech stack and integration topology
- ·Regulatory and compliance posture
- ·Team skill snapshot
Outputs you walk away with
- ·AI Readiness Score — five-dimension index (Strategy, Data, People, Tech, Governance)
- ·Use Case Backlog — 10–30 candidates scored by impact × feasibility
- ·12-month AI Roadmap
- ·Governance & Risk Charter
- ·Executive briefing deck and board-ready summary
Who buys this alone: Organizations that need a credible plan before they spend on AI — especially boards, audit committees, and CEOs about to allocate capital.
Train
Adoption fails on people, not models. We train every layer of the organization with custom curricula built on your data.
Full playbook ›Hide playbook
Build the skills, vocabulary, and confidence the org needs so that AI is adopted, not feared — and so the next phases have an informed audience.
Inputs we work with
- ·Target audiences (executive / functional / technical / board)
- ·Business context, tools in use, current adoption maturity
- ·Real-world data scenarios (anonymized) for custom labs
Outputs you walk away with
- ·C-Suite AI bootcamps — strategy, governance, ROI lens
- ·Functional cohorts — HR, Finance, Sales, Operations, Legal
- ·Technical guild track — prompt engineering, Copilot Studio, agent building, RAG
- ·Custom labs using your own data scenarios
- ·Post-training adoption playbook (this is the bridge to Phase 3)
Who buys this alone: Organizations that already have direction but need their people moving in it. Every cohort ends with a pilot-scoping session — that’s the trojan horse into Phase 3.
Innovate
One high-value use case becomes a working PoC in your environment — with evaluation criteria agreed before we write code.
Full playbook ›Hide playbook
Take one use case from the backlog and ship a real, working prototype in your environment — not a slide deck, not a demo on our laptop.
Inputs we work with
- ·Prioritized use case from the backlog
- ·Sample (or production) data
- ·Success criteria signed off by sponsor
- ·Stack Fit Assessment (cloud LLM vs open-source) — included
Outputs you walk away with
- ·Working PoC running in your Azure / AWS / GCP / on-prem environment
- ·Architecture documentation and decision log
- ·Evaluation report — accuracy, latency, cost, user feedback
- ·Go / no-go business case for Phase 4
- ·Code transferred to your repository on day one
Who buys this alone: Teams ready to stop talking about AI and ship something. Fixed scope, fixed price — no "T&M PoCs" that drift.
Build
The PoC that earned it goes to production — with the MLOps, security review, and knowledge transfer to make it stick.
Full playbook ›Hide playbook
Take what worked in Phase 3 and harden it into a system that runs at scale — with the operations team that will own it after we leave.
Inputs we work with
- ·Validated PoC from Phase 3
- ·Production data access
- ·Integration requirements and SLA targets
- ·Your engineers shadowing from week one
Outputs you walk away with
- ·Production deployment on your subscription
- ·CI/CD, monitoring, and evaluation harness
- ·Operations runbook and on-call playbook
- ·Adoption metrics dashboard
- ·Full knowledge transfer to client team — knowledge transfer is contractual, not optional
Who buys this alone: Organizations with a green-lit pilot ready to go live. We lead; your engineers take over by Go-Live.
Sustain
AI systems drift and costs creep. We stay on a light retainer to keep what we built healthy — and find what to build next.
Full playbook ›Hide playbook
Make sure the AI system that went live in Phase 4 is still working — technically, economically, and organizationally — six months and six years from now.
Inputs we work with
- ·Live system in production
- ·Ongoing use-case pipeline
- ·Model evaluation feedback
Outputs you walk away with
- ·Quarterly health check — model drift, evaluation, governance refresh
- ·Cost & token spend audit with optimization recommendations
- ·Continuous L&D refresh for the operating team
- ·Annual strategy day — where the AI portfolio goes next year
- ·New use-case incubation pipeline (loops back to Phase 1 / 3)
Who buys this alone: Every Build-phase client should be a Sustain-phase client. This is the phase most consultancies skip — and the phase most AI projects die in.
How we keep it honest.
Which phase are you in?
The Gennoor Way is a loop, not a waterfall. Three common starting points.
Starting from zero
You’ve heard about AI, you have budget, but you don’t know where to begin.
Recommended entry
Phase 1 · Diagnose
Full five-phase journey, typically 9–18 months end-to-end.
Strategy already done
You have a roadmap (in-house or from a Big-4 firm) and you’re ready to execute.
Recommended entry
Phase 2 · Train → Phase 3 · Innovate
Skip Diagnose, run a training cohort + a pilot in parallel. ~12 weeks to first working system.
Live model that’s drifting
You shipped an AI system months ago. Adoption is sliding, costs are climbing, accuracy is degrading.
Recommended entry
Phase 5 · Sustain
Quarterly health check + cost audit + L&D refresh. Often loops back into Phase 3 for the next use case.
Not sure which fits? Run the 15-minute diagnostic ›
Questions before you start.
Do I have to start at Phase 1?
No. The Gennoor Way is a loop, not a waterfall. You can enter at any phase based on where your organization actually is. A bank with a roadmap can start at Train or Innovate. A team with a working PoC can start at Build. A live system that’s drifting can start at Sustain.
How is this different from what a Big-4 consultancy offers?
Three differences. First, we deliver the framework end-to-end — diagnostic to deployed agents to long-term sustainment — on one engagement. Big-4 firms typically split this across three different contracts. Second, we publish our price bands; Big-4 firms do not. Third, our delivery is senior-only — no analyst tier between you and the practitioner doing the work.
Who owns the IP — the code, the models, the prompts?
You do. Code is in your repositories from day one. Models, data, and prompts are client-owned. Our reusable frameworks (templates, evaluation harnesses, readiness scoring methodology) remain ours and are listed in the contract.
Can you work on AWS, GCP, or open-source — not just Microsoft?
Yes. Azure is our most-deployed stack because we’re Microsoft Certified Trainers, but we deliver on AWS Bedrock, Google Vertex AI, and self-hosted open-source (Llama, Mistral, Phi, Qwen). Every Innovate-phase engagement starts with a written Stack Fit Assessment comparing cloud LLM vs open-source, so the recommendation is yours, not ours.
Can you work air-gapped or on-premise for regulated workloads?
Yes. Open-source LLMs on private infrastructure — Ollama, vLLM, or Azure ML private endpoints — is one of our reference patterns. Common for government, defense, healthcare, and regulated finance clients who cannot send data to public APIs.
What happens if our team can’t take over after Build?
We extend hypercare and roll into a Sustain retainer. Knowledge transfer is contractual on every Build engagement, but if your team is unable to operate the system after handover, we stay. We have not yet had a client unable to take over — but the path is there if it ever happens.
How long does a typical full transformation take?
For an SMB starting at Phase 1, expect 6–9 months to ship the first production agent and another 3 months to scale to the next two use cases. For an enterprise, expect 12–24 months for the first wave of deployed systems, with Sustain running indefinitely as new use cases emerge.
Can we run our internal team and your team together?
Yes — this is our preferred model from Phase 4 onward. Co-build means we lead, your engineers shadow, and ownership transfers by Go-Live. We also work alongside existing SIs (Accenture, Wipro, TCS) as the AI specialist arm of broader transformations.
Where does the data sit?
In your environment. We build inside your Azure subscription, your AWS account, your on-prem infrastructure — never on Gennoor-owned infrastructure. The only exception is anonymized usage telemetry, which is opt-in.
Do you offer free starting points before we commit?
Yes. The 15-minute AI Readiness Diagnostic is free. Exploratory 30-minute calls are free. Free workshops run quarterly in GCC, India, and East Africa. We charge starting at the Diagnose-phase engagement.
Phase one takes 15 minutes.
The diagnostic is where every engagement begins.