A three-month project is an awkward size. It's too long for a freelancer you found on Friday to wing it, and too short to justify a full-time hire with a recruiting fee, benefits, and a severance conversation at the end. It's also short enough that onboarding takes a real bite. If your new developer spends the first three weeks learning the codebase, you've lost a quarter of the engagement before the real work starts.
So beyond picking who to hire, you need a plan that makes weeks one through three count.
Key Takeaways
- For three months, pick between a freelancer, an agency, and staff augmentation. A full-time hire rarely fits.
- Onboarding can take weeks. Google researchers found the shift to remote onboarding slowed ramp-up by roughly three to six weeks.
- Write the scope and the definition of done before the developer starts, not during week one.
- Plan the handover from day one. The last two weeks belong to documentation and transfer.
Choosing the Model
| Model | Best when | Risks on a 3-month project |
|---|---|---|
| Freelancer | Scope is narrow and well defined; you can review their code | Availability, disappearing mid-project, uneven vetting |
| Agency (fixed scope) | You want a finished deliverable and won't manage daily work | Change orders, less control, handover of an unfamiliar codebase |
| Staff augmentation | The developer works inside your team, on your tools, under your lead | Needs a manager on your side; minimum terms |
| Full-time hire | You'll need the person after the project | Recruiting time can exceed the project itself |
For work that touches your existing codebase and needs daily judgment calls, staff augmentation usually fits best: the developer joins your standups, uses your repo and review process, and you direct the work. For a standalone build you don't want to manage, an agency makes more sense. We compare the first two in more detail in staff augmentation vs freelancers and the agency route in freelancer vs agency to build your app.
A note on minimums, since they matter at this size. Ruzora's engagements start with a 90-day initial commitment, then go month to month with 30 days' notice. That lines up with a three-month project, and it means you aren't locked in past it. Other providers have different minimums; ask before you sign.
The Ramp Problem
When Google studied onboarding, it found that "the shift to remote onboarding resulted in slower ramp-up for engineers on the order of three to six weeks" (DX summary of Google's research). On a 13-week project, even a few weeks of slow start is a big share of the calendar. Microsoft researchers who surveyed 267 new hires, most of them onboarded remotely, found that building connections with the team was one of the hardest parts (Microsoft Research).
You can't make ramp-up disappear, but you can shrink it:
- Have access ready on day one: repo, CI, staging, the ticket tracker, the team chat.
- Assign a buddy who reviews every PR for the first two weeks.
- Give the developer a real, small ticket on day one and a shipped change in week one.
- Choose someone who has built this kind of thing before. Prior experience with your stack cuts ramp more than any onboarding doc.
How to onboard a developer fast has a full checklist.
A Week-by-Week Plan
Before day one: a written scope, a definition of done, the three riskiest parts of the work, and the name of the person who approves decisions.
Weeks 1 to 2: setup, a small fix shipped, pairing on the riskiest part of the project, and a short written plan from the developer that you agree on.
Weeks 3 to 10: the build. Weekly demos, not weekly status reports. If something slips, you see it in the demo.
Weeks 11 to 13: finish, test, document, and hand over. The developer walks your team through the code and writes the runbook. See the software project handover checklist.
A Concrete Version
A 20-person logistics startup needs to replace its hand-rolled carrier integrations with a single abstraction layer before a large customer onboards in April. It's January 5. The in-house team of four is fully booked on the core product.
The CTO writes a two-page scope: five carrier APIs behind one interface, retry and rate-limit handling, and a test suite with recorded responses. She requests a staff-augmentation shortlist on January 5, interviews two senior backend engineers that week, and her pick starts January 26 on a 90-day commitment.
Week one, he ships a small bug fix and reads the three existing integrations. Week two, he proposes the interface; the CTO signs off in a 30-minute call. Weeks three to ten, he moves one carrier over every week or two, demoing each on Friday. In weeks eleven to thirteen he writes the runbook and pairs with an internal engineer on the last carrier so the team can own it. The customer onboards on time in mid-April. The engagement ends at the 90-day mark with 30 days' notice given in March.
The Honest Counterpoint
Three months is often a lie. Projects scoped at three months commonly run longer, which is why the planning fallacy has its own post. If there's a real chance this becomes six or nine months, pick a model that can extend cleanly and a developer you'd want to keep.
And if the work is truly small, say three or four weeks of effort padded into a quarter, three months of a senior engineer is the wrong purchase. A well-vetted freelancer on a fixed scope may be cheaper and faster.
Frequently Asked Questions
Can I hire a full-time employee for a three-month project?
You can, but it's rarely a good fit. Recruiting can take as long as the project, and ending a full-time role after three months is hard on everyone.
How much of a three-month project goes to onboarding?
It depends on your codebase and the developer's experience with your stack. Plan for at least the first two weeks to be mostly ramp-up, and set everything up before day one to keep it there.
What if the project runs over three months?
With staff augmentation, the engagement usually continues month to month. Agree up front how extensions work, whichever model you choose.
The Bottom Line
For a three-month project inside your codebase, a senior augmented engineer with a written scope and a ready-on-day-one setup is usually the fastest route. Ruzora sends a vetted shortlist within 72 hours, and our 90-day initial commitment matches a quarter-long project. Request a shortlist or see how our vetting works.
Roberto Espinoza is CEO of Ruzora, which helps US startups hire pre-vetted senior LATAM engineers, with a vetted shortlist in 72 hours. See available engineers.
