Software Development Engagement Models

Four Models, One Question

Every engagement we run is one of four shapes, and a single question decides between them: who owns the outcome?

If you own it, you are buying capacity. Your leads set priorities, run standup and accept the work; we find, vet and keep the engineers. That is staff augmentation or a dedicated team.

If we own it, you are buying a result: a scope delivered to a date, or a system held to an availability target. That is fixed-scope project delivery or managed IT services.

All four draw on the same bench: 1,000+ engineers across delivery centers in Mexico, Colombia and Bolivia, each cleared through four gates (technical assessment, live coding review, English proficiency and culture fit) from a pool where fewer than 2% of applicants are hired. Sixteen years, 2,500+ projects, 98% client retention, 76 Employee NPS. The model changes who directs the work, not who does it. See the full services range.

How to Choose, and How to Change Your Mind

Match the model to your actual constraint, not to the biggest-sounding option.

• Roadmap and plan, but not enough hands. Staff augmentation. Wrong call if you lack the engineering management bandwidth to direct extra people.
• A product that needs continuity. Dedicated team. Wrong call for a one-off with a hard end date; you would pay for continuity you never use.
• A bounded problem, and you want a date and a number. Fixed-scope delivery. Wrong call if requirements are still moving, because locking scope you cannot specify turns every discovery into a change order.
• Production stability, not feature velocity. Managed IT services. Wrong call if what you really need is to build something new.

You are not locked in. Dedicated teams move into fixed-scope engagements and back again. The most common path is build then operate: a fixed-scope engagement ships the system, managed services runs it. Capacity scales the same way: clients typically start with 2–5 engineers and grow to 20–50+ without re-onboarding. Our engineers compare two of these in staff augmentation vs. software outsourcing, and the wider tradeoffs sit on our software development outsourcing page. Where the work happens is a separate decision from how you engage, and it is covered on our nearshore software development page.

The Four Models Side by Side

	Staff Augmentation	Dedicated Team	Fixed-Scope Delivery	Managed IT Services
Owns the outcome	You	You; we own the team	Unosquare	Unosquare, against an SLA
Manages day to day	Your leads	Your tech lead	Our project lead	Our ops team
You select	Engineers, from 3–5 profiles	The brief, not the people	The scope, not the people	The coverage, not the people
Scoped by	Your backlog, sprint by sprint	Your roadmap	Written spec and acceptance criteria	Service scope and ownership boundaries
Accepted by	Your definition of done	Your definition of done	Acceptance criteria, per phase	SLA targets and incident reports
Flexes by	Adding or dropping engineers	Adjusting team composition	Change order, priced separately	Scaling coverage and environments
Commercial shape	Monthly rate per engineer	Monthly rate for the team	Fixed price for defined scope	Recurring fee against service levels
Carries estimate risk	You	You	Unosquare	Unosquare

Buying Capacity: Augmentation and Dedicated Teams

Staff Augmentation: you manage the engineers

Suits teams with a roadmap, a tech lead and a specific gap: a backlog that will not clear, a skill missing in house, a quarter that needs three more engineers. Does not suit teams without the management bandwidth to direct people, or work you would rather hand over entirely.

Included: individual engineers, vetted and onboarded by us, with the HR, payroll, hardware and compliance logistics behind them, a weekly sync with your tech lead, monthly check-ins, and a replacement guarantee if someone is not the right fit. Managed by: you, like team members, with no separate vendor-management layer. Scoped and accepted: your backlog, your sprints, your code review, your definition of done. Flexes: add or drop engineers with notice. Commercially: a monthly rate per engineer instead of salary, benefits, payroll taxes, recruiting fees and hardware.

You receive 3–5 profiles with resume, GitHub and portfolio, typically within 5–10 business days for standard stacks; you run the technical assessment, we run culture fit and references. Two to four weeks from request to productive. See staff augmentation services, or hire developers by stack.

Dedicated Teams: we shape the team, you own the product

Suits products that need continuity: a team that learns your codebase, your tech debt and why the decisions were made, and gets faster because it stayed. Does not suit a single bounded piece of work with a hard end date.

Included: a team shaped around your brief (seniority mix, expertise, the roles the work actually needs) working exclusively on your product, with composition optimized as your roadmap moves. Managed by: your tech lead day to day; we own who is on the team and handle replacements proactively, without putting you back through sourcing and interviews. Scoped and accepted: your roadmap, your definition of done. Flexes: scale up after a raise or down after a launch, with no internal hiring or layoffs. Commercially: a monthly rate for the team.

Nonstop Health has run this way for seven years: four core portals for brokers, clients, members and admins, covered by 1 technical lead, 1 release manager, 1 business analyst, 3 IT ops engineers, 18 software engineers and 4 QA engineers, with an embedded recruitment program averaging a 14-day time-to-fill. Read the case study, or see dedicated development teams.

Fixed-Scope Project Delivery

Suits a bounded problem you can specify (a rebuild, a migration, an MVP, an integration) where you want a date and a number more than you want a team to manage. Does not suit requirements that are still moving, or open-ended product development.

Included: discovery, architecture, build, QA and launch as defined phases with deliverables; production-ready code with testing, security review and documentation; and a post-launch support window. After it closes you own the system outright, with run books and architectural guides. Managed by: us, with one project lead, one engineering team and no churn. Scoped and accepted: requirements, stack and acceptance criteria documented before work starts, with each phase accepted against them. Flexes: a new requirement mid-project becomes a written change order, priced separately, rather than quiet scope creep. Commercially: one fixed price, known before work begins. If we underestimate, we absorb it.

GTreasury used both halves of this model. An eight-week engagement produced a transition strategy, a modernization roadmap and a census of hundreds of legacy Delphi plugins (read it). The build that followed rebuilt 400+ product screens in under eight months, twice as fast as planned and under budget (read it). More on outcome-based delivery.

Managed IT Services

Suits teams whose constraint is production, not roadmap: engineers losing their week to alerts, patching and 2am debugging, or a system someone else built that now has to stay up. Does not suit a team whose real need is to build something new.

Included: infrastructure and compute across AWS, Azure and GCP; databases; monitoring; security and compliance work including vulnerability scanning, patching and secrets rotation; CI/CD and deployment; and 24/7 incident response with on-call engineers, playbooks and post-mortems. Plus the proactive half: capacity forecasting, performance optimization and automation of backups, deployments and credential rotation. Managed by: us. Scoped and accepted: service scope and ownership boundaries documented up front, measured against SLA targets, with co-managed available if you want to keep some areas. Flexes: coverage scales with your environments and stack. Commercially: a recurring fee against service levels, fully loaded, with no per-incident charges.

We can take over systems we did not build, through codebase walkthrough, architecture review and run book development. Axos Bank has run this way for seven years, raising reliability from 98.7% at onboarding to 99.9% uptime while unifying the technology of five acquisitions, with 9 infrastructure engineers and 8 reliability engineers on 24/7 monitoring. Read the case study, see our financial services work, or go to managed IT services.

What You Are Actually Signing

We do not publish a rate card, because a rate without a scope is marketing. What we can be precise about is the structure, and the structure decides who carries the risk.

Three commercial shapes. Capacity models bill like time and materials: labor at agreed rates which, as FAR 16.601 puts it, are "fixed hourly rates that include wages, overhead, general and administrative expenses, and profit." The same rule is candid that this shape "provides no positive profit incentive to the contractor for cost control or labor efficiency", precisely why you keep management control in augmentation and dedicated teams. Fixed-scope delivery inverts that: a firm-fixed-price arrangement, per FAR 16.202-1, "places upon the contractor maximum risk and full responsibility for all costs," and is only sane against "reasonably definite functional or detailed specifications", which is why the spec comes first. Managed services is the third shape: a recurring fee against service levels rather than hours or deliverables.

What moves the number: seniority mix, how scarce the stack is, the shape of the team, engagement length, timezone overlap, and compliance burden: audit trails and validated environments are engineering work, not paperwork. See the stacks we cover.

What to expect in a quote: team shape, timeline and cost, scoped before you commit. Rates are fully loaded (no separate line for benefits, overhead or infrastructure), so the honest comparison is not our monthly rate against a US salary but against the fully loaded internal cost of the same seat. Agreements are structured to fit your project duration and fiscal planning rather than a standard term.

Ownership. Settle IP in the contract, not by assumption. The US Copyright Office's Circular 30 states that where a work is made for hire, "the employer or the party that specially ordered or commissioned that work is the initial owner of the copyright." But it is strict about when commissioned work qualifies: the work "must fall within one of the nine categories" the statute lists, and software is not among them. So for outsourced development the phrase "work made for hire" does not carry the weight on its own; express written assignment does. Put it in writing before onboarding.

Engagement Model FAQ

Still Deciding?

Tell us what you are building and what is in the way. We will come back with the model, the team shape, the timeline and the cost, before you commit to anything.

Contact Us