DEBCOR Engineering®

DEBCOR AI Packages

AI Ready in 30 Days

AI Strategy & Readiness Assessment

In three to four weeks: a ranked list of which AI projects will pay off on your systems, which won't, what order to run them in, and what each will cost.

Every AI project you ship inherits an inference bill. This assessment gives you the ranked project list plus the tiered token strategy — frontier reasoning, efficient extraction, and where RAG kills the LLM call entirely — so the cost forecasts actually hold.

Timeline3–4 weeks to a decision-ready roadmap to production

What's inside

Everything the package ships with.

A ranked list of AI use cases across your business — scored on business impact × feasibility × infrastructure fit, not vendor pitch decks

A profiling report on your top data domains — what's ready for AI today, what needs cleaning first, what to leave alone

A tiered token strategy — which classes of workload run on frontier reasoning models, which run on high-efficiency low-cost models, and where RAG against your own data replaces the LLM call entirely

An inference cost model per use case — token forecasts, model-routing rules, ceiling scenarios, and the crossover where hosted-vs-cloud economics flip

A governance and risk framework — human-in-the-loop rules, audit trail requirements, PII handling, and vendor lock-in mitigation

A 12–24 month sequenced roadmap with a 90-day quick-win recommendation — so the board sees momentum in a quarter, not a year

Sizing & published pricing

Three sizes. One architecture.

Scope compresses down, not features. S-tier is playbook-driven — a single senior engineer delivers against pre-built accelerators inside a fixed configuration boundary. M-tier extends the same systems into landscapes with multiple sources, custom rules, and real integration complexity. L-tier is the full enterprise engagement: dedicated environments, SOC 2 wrap, roadmap co-development, and — for SAP customers — clean-core BTP integration. Same architecture at every size.

Starter

Band

≤ 500 employees · one system, one team

Timeline

2–3 weeks

Published price

from $9k

Scope at this tier

Focused readiness — one business unit, up to 3 use cases scored, single-domain token strategy.

Mid-market

Band

500–5,000 employees · multiple sources or rules

Timeline

3–4 weeks

Published price

$28–50k

Scope at this tier

Enterprise readiness — cross-BU use case portfolio, tiered token strategy across all workloads, full governance framework.

Large

Band

5,000+ employees · enterprise scope, SOC 2, roadmap co-development

Timeline

4–6 weeks

Published price

$65–110k

Scope at this tier

Multi-entity readiness (PE portfolio, holding company, global) — per-entity token strategies with cross-entity model sourcing, board-ready deliverable.

Published bands are directional. Final commercial terms are per-client SOW. Managed operations are quoted separately.

Timeline

3–4 weeks to a decision-ready roadmap from scope to live.

Every package follows the same four-phase delivery. What changes at each tier is the count of sources, rules, and integration surface — not the shape of the work.

  1. Discover

    Week 1

    Stakeholder interviews, source-system inventory, data pull, current-state audit.

  2. Profile & Score

    Weeks 2–3

    Data-quality profiling, use-case ranking, tiered token strategy, gap analysis.

  3. Roadmap

    Weeks 3–4

    Sequenced 12–24 month plan + 90-day quick-win recommendation + per-use-case inference cost model.

  4. Readout

    Week 4

    Executive readout, decision-ready deliverable, follow-on scoping call.

For SAP customers · Bridge to SAP-native

For SAP customers, the assessment adds RISE entitlement and Joule Studio 2.0 readiness alongside the general AI plan. The tiered token strategy accounts for BTP-hosted model routing, embedded Joule agent costs, and where non-SAP model routing keeps token spend down. See how DEBCOR extends this into SAP →

Package FAQ

Common questions.

Token strategy is the discipline of matching every AI workload to the right tier of model — frontier reasoning for genuinely hard problems, high-efficiency low-cost models for the volume work, and RAG (retrieval against your own data) where no LLM call is needed at all. Without it, you either burn cash routing everything to a frontier model, or you cripple quality by putting everything on the smallest tier. The assessment produces per-use-case routing rules and a forecast so you know what each project will actually cost at scale — not just what a vendor's per-million-token headline suggests.

We benchmark against your specific workloads, not generic leaderboards. For every use case in the ranked plan, we identify which model tier (frontier, mid, small) handles it, which vendor's tier fits best on cost / latency / governance / lock-in, and where model routing lets you swap over time. The output is a matrix, not a religion — different workloads get different models.

DEBCOR is an Anthropic Preferred Services Partner, so Claude is often the strongest fit for governance-heavy enterprise work. But the assessment is workload-first — if Gemini is stronger for a specific extraction task or GPT is cheaper for a specific bulk-classification job, that's what the roadmap will say. You get the ranked plan and the rationale, not a vendor pitch.

Those are treated as fixed inputs. The assessment plans around them — where they cover the use case well, we use them; where they don't (multimodal ingestion, custom agents, private RAG), the roadmap slots in the right model tier and routing strategy alongside them. You don't have to unwind an existing commitment to get value.

Yes if you run SAP. RISE entitlement audit, Joule Studio 2.0 readiness, BTP-hosted model routing options, and where non-SAP model routing keeps token spend down alongside SAP-native agents — all included from the M tier for SAP customers, with L tier extending to multi-entity SAP landscapes.

Scope this package

Fixed scope, fixed price. Tell us the document types and the systems, and we'll come back with a one-page proposal.

Start the assessment

Not sure this is the one?

Book a 30-minute fit call with a senior architect. We'll tell you if this is the right package, a different one, or nothing at all.

Book a 30-minute fit call