AiDOOS Glossary

Outcome-based delivery is the model where the customer pays the vendor for shipped, accepted business outcomes rather than for engineer hours, named headcount, or fixed-bid scope estimates. The category emerged in response to the structural failures of T&M billing (vendor profits when work takes longer), per-FTE subscription (customer pays for redundancy they didn't ask for), and fixed-bid SOWs (scope changes trigger contracting cycles).

By 2026 the phrase is widespread — Big-IT services firms, freelance marketplaces, and EOR platforms all market "outcome-based" delivery. AiDOOS is the operating model that makes it structurally true: Delivery Units as the pricing primitive, Virtual Delivery Centers as the execution infrastructure. The customer cannot pay for unshipped work because the engine cannot bill it.

Real Day 0 is the AiDOOS framing for the calendar reality of outcome-based delivery. When AiDOOS commits to shipping against milestones, Day 0 is the day scope is aligned — first commit follows in 5-10 business days. When a Big-IT consultancy commits to the same, their Day 0 follows 60-180 days of RFP issuance, multi-vendor selection, MSA negotiation, partner-level chartering, and account-staffing cycles.

AiDOOS collapses that pre-engagement period to zero structurally: pre-vetted bench (no staffing lag), pre-published standardized MSA + DPA (no legal-cycle lag), AI-matched pod composition (no committee selection lag), and self-service contracting at the Starter tier (no enterprise sales cycle). Real Day 0 is what outcome-based delivery means at the calendar level — others claim 14-day onboarding, only AiDOOS makes Day 0 actually Day 0.

Structurally outcome-based is the brand differentiator that distinguishes AiDOOS from vendors who claim outcome-based delivery while structurally billing hourly. The four mechanics that make AiDOOS structurally outcome-based:

1. DUs only consume against accepted milestones. No DU consumes if the work didn't ship and acceptance criteria weren't met.
2. Unused DUs are refundable. Customer recovers budget for unallocated capacity at the rate paid, no questions asked.
3. Re-delivery on acceptance miss is platform-funded. The customer never pays twice for the same outcome.
4. Pricing is published, not negotiated. $200/$167/$160/<$140 per DU across four tiers, identical to all customers at the same DU count.

These mechanics cannot be retrofitted onto hourly billing. They are properties of the operating model.

The delivery manager is the full-time, embedded role that distinguishes a VDC engagement from staff augmentation. The DM owns sprint cadence, code-review SLAs, milestone reporting, customer communication, and pod-composition decisions. Every AiDOOS pod includes one DM. The customer's engineering manager spends 1-2 hours per week on the engagement instead of 8-10, because the DM absorbs the operational management layer.

The DU Dictionary is the reference catalog AiDOOS uses to convert real-world work to Delivery Units. AI-assisted estimation matches a story description against the catalog; the calibration board adjudicates unusual cases; continuous learning from actual-vs-estimated outcomes refines the calibration over time. The Dictionary is AiDOOS's core IP and competitive moat — competitors can copy the pricing page in a day; they cannot copy years of calibration data.

A pod is the delivery unit inside a VDC engagement — typically 3-7 specialists (engineers, designers, QA) plus a full-time delivery manager. Composition adapts to the engagement scope: a frontend-heavy product build pairs differently from a data-platform modernization. Pods scale up, down, or pause without contract amendments — the platform handles internal staffing changes invisibly to the customer.

AiDOOS's talent pool is a global, continuously-vetted network of 100,000+ technology professionals matched to engagements via AI based on stack expertise, sector experience, seniority, availability, and historical delivery performance. Vetting includes portfolio review, AI-driven technical assessment, and live engineering interviews. Continuous performance signals from delivered work feed back into platform-level talent ranking.

The verification framework is the platform-level system that adjudicates delivery quality regardless of who or what produced the output (human or AI agent). It includes acceptance criteria validation, code review gates, customer feedback loops, and continuous performance scoring that feeds into talent ranking. The framework is what makes the re-delivery guarantee and DU pricing structurally reliable.

A Virtual Delivery Center is AiDOOS's core execution unit — a pre-assembled pod of vetted technology professionals plus an embedded delivery manager and the tooling to ship outcomes against a milestone schedule. A VDC differs from staff augmentation (which sells worker-hours), from outsourcing (which sells fixed teams on long contracts), and from freelance marketplaces (which sell access to individuals). The VDC sells delivery — pre-vetted talent, embedded management, and accountability for shipped output.

The $/DU rate is the single per-DU price the customer sees, varying only by tier: $200/DU at Starter, $167/DU at Small, $160/DU at Scale, and below $140/DU at Enterprise. Internally, AiDOOS multiplies a base unit by factors capturing technology, seniority, scarcity, and complexity to determine how many DUs a piece of work consumes — but those factors are never shown as customer line items. The customer thinks in DUs; the engine handles the rest.

A Delivery Unit is the universal output-based pricing primitive AiDOOS uses across all engagements. One DU is a standardized measure of cognitive output, calibrated against the DU Dictionary. As a benchmark only: 1 DU ≈ 4 hours of mid-level Java engineering output — but the hours are a benchmark, not a billing unit. A senior architect shipping 1 DU in 1 hour and a junior engineer shipping 1 DU in 8 hours both produce 1 DU. The customer pays the same regardless of who shipped or how long it took.

A credit pack is a pre-purchased bundle of DUs that the customer draws against as work ships. The four pack tiers — Starter (10 DUs / $2K / 90 days), Small (60 DUs / $10K / 6 months), Scale (250 DUs / $40K / 12 months), and Enterprise (custom) — represent the same engine pricing expressed at different volume commitments. Unused DUs in a pack roll forward when topped up, or can be refunded at any time within the validity window.

Pre-flight estimation is the customer-facing transparency mechanism that shows DU consumption before work begins. The Instant Proposal generates an estimate by matching the work description against the DU Dictionary; the customer sees the estimated DU count, the historical accuracy band of similar estimates, and the breakdown of what drove the size. Estimates can be revised during scoping; once accepted, the platform commits talent against the estimate.

The Project flow is the entry path for customers who don't want to commit to a credit pack up front. The Instant Proposal sizes the work in DUs; the resulting DU count determines which tier band applies (1-30 DUs at Starter rate, 31-150 at Small rate, 151-499 at Scale rate, 500+ at Enterprise). The customer pays the tier rate for every DU. There is no on-demand premium — Project flow and pre-purchased packs share the same rate card at the same DU count.

The re-delivery guarantee is one of three trust mechanisms wrapped around DU pricing. If delivered work fails the customer's documented acceptance criteria, AiDOOS re-delivers without consuming additional DUs from the customer's wallet. The platform pays the talent for the re-do; the customer's economics are bounded — they cannot pay twice for the same outcome. This shifts quality risk from the customer to the platform.

Refundable unused DUs is the trust mechanism that competitors structurally cannot match. DUs sitting in the customer's wallet — not yet allocated to active in-progress work — can be refunded at the rate paid, no questions asked, to the original payment method. This costs AiDOOS nothing on the unconsumed side (no talent has been paid against those DUs), but provides a structural risk bound the customer cannot get from per-seat subscription contracts (Toptal, Deel) or hourly engagements (staff aug, Big-IT).

Trust posture is AiDOOS's term for the bundle of three buyer-protection mechanisms that compose the total risk profile of DU pricing: pre-flight estimation (you see the cost before you authorize), re-delivery on acceptance miss (you don't pay twice), and refundable unused DUs (you can recover the unconsumed portion at any time). Together, these remove essentially every buyer objection that competitors leave on the table.

The validity window is the period during which DUs in a credit pack can be consumed against shipped work or refunded if unused. Starter packs run 90 days, Small packs 6 months, Scale packs 12 months, Enterprise packs are custom. Within the window, unused DUs can be refunded at the rate paid; topping up before expiry rolls remaining DUs forward into the new pack. After expiry, unused DUs are forfeited unless an Enterprise contract specifies otherwise.

Acceptance criteria are the explicit, documented conditions under which delivered work counts as accepted. Tight acceptance criteria are the customer-side discipline that makes DU pricing work — they're the basis for milestone sign-off, the trigger for DU consumption, and the gate for the re-delivery guarantee. Vague acceptance criteria lead to disputes; sharp ones make the engagement run cleanly.

Capability-based engagements are multi-year arrangements where AiDOOS provides an ongoing engineering function — the data-engineering capability, the MLOps capability, the platform-reliability capability. Capacity is sized for the function's ongoing needs; pod members may rotate but the function persists. Typically uses Enterprise tier with custom DU commitment and quarterly business reviews.

An engagement is the scoped relationship — typically defined by a Statement of Work — under which AiDOOS delivers outcomes for a customer. Engagements vary by shape: project-based (bounded scope, fixed milestone schedule), retainer-based (ongoing capacity over time), or capability-based (long-running embedded function like data-platform engineering). The DU pricing primitive applies across all shapes.

A milestone is a defined, acceptable unit of shipped work — typically 2-4 weeks of pod output. Each milestone has documented acceptance criteria; DU consumption is tied to milestone acceptance. Smaller milestones add ceremony overhead; larger ones reduce predictability. Most engagements settle on 3-week milestones as the sweet spot.

Pod composition is the platform's choice of role mix and seniority distribution for a given engagement. A frontend-heavy product build pairs 2 frontend specialists with a designer and a backend engineer; a data-platform engagement pairs 2 data engineers with a platform engineer. Composition is the platform's responsibility — the customer specifies outcomes; AiDOOS sizes the pod. Customer DU economics don't change whether the platform staffs 4 or 6 engineers.

Project-based engagements are the most common shape — a defined scope, milestone schedule, and bounded duration (typically 3-9 months). Used for new product builds, modernization, transformation programs, and time-bounded integration work. DU consumption tracks against shipped, accepted milestones. The pod rolls off at engagement end (or transitions to a follow-up).

Retainer engagements reserve pod capacity for ongoing, less-defined work — maintenance, feature acceleration, R&D, embedded engineering. Typically uses Scale or Enterprise tier with a steady DU consumption rate over time. Customer-side product management actively feeds work to the pod; underutilization is the customer's failure to manage, not the platform's.

Bench tax is the cost the customer absorbs when paying for engineer time that doesn't go into shipped work — ramp time, between-milestone gaps, scope-clarification cycles. In hourly billing and per-FTE subscription, the customer pays it. In AiDOOS DU pricing, the platform absorbs it — the customer pays only for shipped, accepted DUs. This is the unit-economics fix that lets a delivery platform overstaff for redundancy without punishing buyers for the redundancy.

Pyramid staffing is the consulting industry's margin-optimization pattern: a few senior people on top, layers of mid-level managers, and many junior associates doing the actual work — with the entire pyramid billed at senior-rate equivalents. Common at Big-4, Accenture, TCS, and similar firms. AiDOOS pricing eliminates the pyramid markup because customers pay per DU shipped regardless of who staffs the pod.

Ramp tax is the velocity gap between engagement start and steady-state shipping — typically 60-90 days for a meaningful codebase. During ramp, engineers are learning the codebase, integrating with tooling, and calibrating to the customer's operational practices. Hourly contracts let ramp tax accumulate on the customer's invoice. AiDOOS DU pricing means the platform absorbs the slow first weeks; the customer pays for shipped output, not learning time.

Total Cost of Delivery is the framework that normalizes delivery costs across pricing models. It includes the headline rate, plus hidden costs: management overhead, ramp tax, bench tax, knowledge attrition, coordination costs, scope-creep absorption. Compared against TCD, hourly engagements that look cheaper on rate often run more expensive on shipped-feature basis. The TCD framework is procurement-grade; rate-card-only comparison is procurement-naive.

Vendor lock-in is the structural friction — long contracts, custom platforms, knowledge concentrated in vendor staff, complex change-order procedures — that makes switching delivery vendors expensive or impossible. A deliberate competitive moat for traditional outsourcing. AiDOOS deliberately removes lock-in: short validity windows, refundable unused DUs, customer-owned IP, knowledge in artifacts not heads, and pod composition flexible at the platform layer.

BOT is the model where a vendor stands up an offshore delivery team, operates it for a defined period, then transfers ownership to the customer. Used by enterprises that want to eventually own their offshore capability but lack the speed or local knowledge to set it up directly. Strengths: eventual ownership. Weaknesses: long timeline, contractual complexity. AiDOOS engagements can be configured for transfer-readiness if the customer ultimately wants to own the function in-house.

A COE is a centralized team — usually in-house — focused on a specific domain such as data engineering, ML, or platform reliability. Distributes specialist capability across multiple product teams. AiDOOS capability-based engagements act as instant, on-demand COEs without the headcount investment.

An EOR (Deel, Remote, Oyster) is the legal and payroll wrapper that lets an enterprise hire individual contractors in jurisdictions where it doesn't have legal entity coverage. The EOR handles employment compliance; the customer still owns hiring, management, and delivery. EOR is complementary to AiDOOS, not a substitute — AiDOOS owns delivery; EOR is one mechanism for the legal wrapper around individual workers.

A GCC is a captive offshore center the enterprise owns directly — its own legal entity, real estate, hiring, and operations. Strengths: full control, knowledge retention. Weaknesses: 12-24 month setup, hundreds-of-employee scale required to amortize overhead, executive attention required. A VDC gives enterprises the same capability instantly, without setting up legal entities or hiring hundreds of employees.

An ODC is a dedicated offshore team for one client, typically with its own facility and named staff. Strengths: dedicated capacity, deep customer knowledge over time. Weaknesses: high setup cost, slow ramp, talent attrition risk, geography-bound. A VDC provides the same dedicated capacity without the infrastructure, real estate, or named-headcount commitment.

Offshoring is the practice of locating delivery work in lower-cost geographies. Orthogonal to the choice of delivery model: an engagement can be offshored under any of staff aug, outsourcing, captive ODC, or VDC. AiDOOS's talent pool is global by construction; pods are AI-matched on capability, seniority, and timezone fit rather than geography per se.

Outsourcing typically means a multi-quarter engagement with a fixed team at a vendor, billed via fixed-bid SOW or T&M. Strengths: cost arbitrage, single counterparty. Weaknesses: fixed team rigidity, slow scope adaptation, long wind-downs, vendor lock-in. AiDOOS provides the cost economics of outsourcing with the flexibility of platform-managed delivery and DU pricing.

Staff augmentation is the model where a vendor places individual contractors with the customer and bills hourly or monthly. The customer is responsible for managing the work, running standups, reviewing code, and handling underperformance. AiDOOS replaces staff augmentation by providing fully-managed delivery: the platform handles management, the customer reviews shipped outcomes.