How to onboard offshore developers in your first week

To onboard offshore developers in your first week, write everything down before they start, record every handoff as Loom video, and ship one merged PR by end of day two. Offshore means a 4-plus-hour timezone gap with your home team. That gap kills any onboarding plan that relies on a synchronous huddle, a "tap the senior on Slack" culture, or a Monday kickoff call as the foundation. The fix is async-first by default, not by accident.

Most offshore onboarding fails not because the engineer is weak, but because the company onboards them like they're nearshore. They schedule a 9am EST kickoff for a developer in Manila (10pm local), they send a Notion link with no audio walkthrough, and they wait for the developer to "ramp" while burning the first three days on payment-setup paperwork. By Friday, no commit. By week two, the founder is questioning the hire.

This post is the offshore-specific playbook: what to do on day zero (before they badge in), what day one through five look like when the team is split between San Francisco and Bangalore, and the operational decisions (Wise vs Deel, VPN posture, recorded handoffs, follow-the-sun PR review) that determine whether you ship or stall.

If your timezone gap is under 4 hours, read our broader guide to onboarding remote developers instead. That post covers the general remote-first playbook. This one is for the harder case: when your engineer is asleep while you work, and vice versa.

Offshore vs nearshore vs onshore: pick your onboarding mode

The right onboarding plan depends on how much synchronous overlap you actually have. Confusing these three is the most common onboarding failure mode.

Approach	Timezone gap	Typical pairings	Daily sync window	Onboarding mode
Onshore	0-3 hours	NYC to LA, London to Lisbon	4-6 hours overlap	Synchronous, screen-share kickoff
Nearshore	3-5 hours	NYC to Buenos Aires, London to Cairo	2-4 hours overlap	Hybrid, daily live standup possible
Offshore	6-12 hours	SF to Bangalore, NYC to Manila	0-2 hours overlap	Async-first, recorded everything

Onshore onboarding tolerates a sloppy README because the new engineer can DM their lead at any time. Offshore onboarding does not. Every minute of fuzziness in your docs compounds across a 12-hour delay. Ask a clarifying question at 6pm SF time, the answer arrives 18 hours later, you have lost a full work day.

The frame shift: in offshore mode, written context is the product. The codebase is secondary. If your written context is good, the codebase explores itself.

Day zero: the 72-hour pre-flight checklist

Most "first week" plans actually need to start 72 hours before the engineer's first day. The reason is mechanical: contractor paperwork, payment rails, and access provisioning have hard latencies that no amount of urgency removes.

Do these before they badge in:

• Contract signed and counter-signed via Deel, Remote.com, or DocuSign. Allow 24-48 hours for the engineer to read it in their timezone.
• Payment rail set up. Wise for direct bank transfers in 80-plus countries, Deel for managed contractor payments with compliance handled, or Stripe Connect if you already use it for marketplace flows. Pick one before they start; switching after they invoice is painful.
• Hardware delivered or stipend wired. A $1,500 one-time laptop stipend wired via Wise day-zero beats waiting for a MacBook to clear customs in two weeks.
• All accounts provisioned with the engineer's personal email first. Linear, GitHub, Slack, 1Password, your cloud console. Use their personal email because the company email may take 48 hours of IT setup; you can transfer access later.
• A written 30-minute Loom from the founder or hiring manager. Why the company exists, what the engineer is going to ship in their first sprint, who else is on the team. Recorded once, watched on their Sunday evening before they start.

If any of those five items is unchecked on day one, the engineer cannot ship. They will instead spend their first three days on logistics. That is the most expensive thing you can do.

Day one: the async-first kickoff

Forget the synchronous kickoff call. For offshore, day one runs entirely async, in the engineer's local timezone, with a written brief and a recorded walkthrough waiting in their inbox when they wake up.

Your day-zero artifact should include:

1. A 5-minute "what you're doing this sprint" Loom from the engineer they will work with most. Code shown, repo navigated, sample task pointed at.
2. A written first-PR brief: the file to change, the test to make pass, the expected diff size (under 100 lines). Make this trivial. Day one is for proving the loop works, not heroics.
3. A link to your team's async standup channel with the format pinned. Three things daily: shipped yesterday, shipping today, blocked on.
4. A link to a shared Notion or Linear doc titled "Questions for [their manager]." Encourage them to dump questions there instead of DMing. Questions get answered in batches when timezones overlap, not pinged one at a time.

The engineer's day one is now: watch the Loom, clone the repo, read the brief, write the first PR, post async standup. Total synchronous time required: zero minutes. If they ship the trivial PR by their end of day, you have a working loop. If not, you have a payment, access, or context problem to diagnose, not a "they didn't understand the call" problem.

For the written brief itself, Loom for engineering communication covers the recording discipline in more depth. The short version: 2 to 5 minutes per video, auto-transcript on, dropped in the PR description and the welcome doc.

Day two to three: follow the sun PR review

This is where offshore onboarding either clicks or stalls. The pattern that works is "follow the sun" PR review: the engineer pushes a PR at end of their day, the home team reviews while they sleep, the engineer wakes up to actionable comments and can iterate immediately.

The mechanics that make this work:

• PR template enforces context. Title, what changed, why, how tested, screenshots/Loom for UI work, linked Linear ticket. No 1-line PR descriptions.
• Reviewer SLA of 8 working hours, posted publicly. Slack channel pinned with the SLA. Founder enforces it for the first month.
• No "approval optional" reviewers. Every PR has exactly one required reviewer assigned at open time, not a CODEOWNERS lottery.
• CI must pass before review. Reviewer's time is too expensive to spend catching broken tests. Vercel preview deploys for frontend, Supabase or Neon branch databases for backend, full Playwright run on every PR.
• A 30-second "what I'm working on" Loom attached to any PR over 200 lines. Voice context compresses 15 minutes of code-reading into 30 seconds.

Every engineer on Cadence is AI-native by default, vetted on Cursor and Claude Code fluency before they unlock bookings. That matters here: an AI-native offshore engineer can review their own PR with Claude, catch most reviewer-noise issues before opening, and ship cleaner first drafts. Median time to first commit on Cadence offshore bookings is 27 hours from booking confirmation.

Day three to five: written everything discipline

By mid-week one, the cracks show. Information that should be written is still verbal. Decisions made in DMs vanish. The offshore engineer, who was never in the DM, asks a question that was already answered three days ago in a 1:1.

The fix is operational, not cultural. Three rules:

1. No decisions in DMs. Direct messages are for "I'm logging off" and "can you check this when you're up." All decisions move to a public Slack channel or a Linear comment. If a decision happened in a DM by accident, the first person to notice copies it to the right place.
2. Every meeting has a written outcome. A 5-line summary in the Linear ticket or Notion page within an hour of the meeting ending. No summary, no meeting happened.
3. Loom every handoff. A handoff is any moment one person stops working on something and another picks it up. The handing-off engineer records a 60-second voice memo or screen recording explaining state. The receiving engineer watches it when they badge in.

The async standup, written correctly, makes most of this enforce itself. Our guide to async standups covers the format and the failure modes in depth. The short version: three lines per engineer per day, posted on a deadline in the engineer's local timezone, threaded for follow-up.

VPN, security, and access posture for offshore

Offshore raises legitimate security questions that nearshore mostly does not. Most are solved with defaults; a few need explicit decisions.

The defaults to ship before day one:

• Hardware MFA via YubiKey or platform passkeys for GitHub, your cloud console, and your password manager. SMS-based MFA is not acceptable for offshore access; SIM swap risk is real in several countries.
• 1Password or Bitwarden with shared vaults scoped to project, not team-wide.
• GitHub branch protection with required reviews and required CI. No direct pushes to main, no exceptions.
• Read-only access to production by default. Write access is requested per-incident in a logged Slack channel, granted via your access tool (Teleport, AWS SSO, or even a manual Vanta workflow).
• VPN only if your application requires it. Most modern SaaS does not. Tailscale gives you per-user zero-trust access to staging in 15 minutes if you do need it. Old-school IPsec VPNs add a week of setup and a recurring connectivity headache; skip them unless compliance forces the issue.

The compliance question (SOC 2, HIPAA, GDPR) is usually less about offshore vs onshore and more about whether the engineer has access to regulated data. Treat the access decision per-engineer, per-system, not per-location. A contractor in Bangalore with read-only access to non-PII staging data is no different from a contractor in Austin with the same access.

Payment infrastructure: Wise vs Deel vs Stripe Connect

The day-one experience for the engineer is mostly about whether they get paid on time, in their local currency, without losing 6% to FX spreads. Pick the wrong rail and you will spend hours every month doing reconciliation manually.

Tool	Best for	Cost	Setup time	Compliance handled
Wise Business	Direct contractor payments, 80-plus countries	0.4-1% FX margin	Same day	No, you handle 1099/contracts
Deel	Managed contractor + EOR	$49/mo per contractor	1-3 days	Yes, contracts and tax forms
Stripe Connect	Already on Stripe, marketplace flow	0.25% per transfer	1-2 weeks	Partial, you still issue contracts
Cadence	Vetted offshore engineers with payment handled	$500-2,000/week, all-in	2 minutes	Yes, contract and payment included

If you have 1 to 3 offshore contractors and you are billing them as independent contractors, Wise is the right call. The FX is honest, the setup is same-day, the engineer sees money in their local bank within 1 to 2 business days.

If you have 5-plus contractors or you need EOR for tax compliance in countries like Brazil or Argentina, Deel is worth the $49/month per seat. The contract templates, tax forms, and compliance audit trail save real time.

Cadence handles this end to end. The engineer is paid Friday for the week's work, you get one invoice from Cadence, and the contractor relationship sits on Cadence's books, not yours. For founders who do not want to set up Wise or Deel just to try one offshore engineer, the 48-hour free trial is the lowest-friction starting point.

What to do this week

Pick one offshore engineer to onboard. Run the 72-hour pre-flight. Ship the day-one brief. Record the kickoff Loom. Enforce the follow-the-sun review SLA for two weeks. At the end of week two, look at the data: PRs merged, reviewer turnaround time, async standup adherence, blocker resolution time.

If the data is healthy, your offshore onboarding works and you can scale to a second engineer. If the data is broken, the diagnosis is almost always one of three things: payment is late (engineer disengages), context is verbal (engineer is blocked on questions), or PR review is slow (engineer waits half the day to learn their PR is wrong).

If you do not want to build all of this from scratch, Cadence ships offshore engineers with the full async-first onboarding kit pre-built: contract, payment via Wise, Loom-first handoff discipline, daily ratings, weekly retros. Find your offshore engineer in 2 minutes and you skip the pre-flight entirely.

How offshore changes performance management

A note for week two and beyond. Once the engineer is shipping, the management mode shifts from onboarding to performance. Offshore performance management is its own discipline; the short version is that signals shift from observation (you cannot see anyone work) to artifacts (PRs, Loom recordings, Linear status, written status updates).

Our deeper write-up on performance reviews for remote engineers covers the cadence and the artifact-based scoring model. For offshore specifically, two adjustments: weekly written status is non-negotiable (Loom video plus 3 bullets), and the quarterly review is built entirely from the artifact trail, not from "vibe."

The compensation question for offshore (do you pay Bangalore rates or San Francisco rates) is upstream of onboarding but worth flagging now. How to compensate remote engineers fairly across geos walks the three models. Pick before you make the offer; renegotiating compensation in week three after the engineer has discovered their teammates earn 4x is a worse outcome than picking a fair rate up front.

Onboarding an offshore engineer this week? Cadence ships vetted offshore engineers with contract, payment, and async-first onboarding handled. Junior at $500/week, Mid at $1,000, Senior at $1,500, Lead at $2,000, all weekly with a 48-hour free trial. Find your engineer in 2 minutes.

FAQ

How long should offshore developer onboarding take?

The first merged PR should ship inside 48 hours of the engineer's start date. Full ramp to owning a feature end to end is typically 2 to 3 weeks for a Mid, 1 to 2 weeks for a Senior. If the first PR has not shipped by end of day two, the blocker is operational (access, payment, context), not the engineer.

What is the best timezone overlap for offshore developers?

2 to 4 hours of synchronous overlap per day is ideal. For SF-based teams, that means engineers in Europe (mornings overlap) or Latin America (full overlap, but technically nearshore). For NYC teams, India is workable with the engineer taking late-evening calls 2 to 3 times a week. Less than 2 hours of overlap is workable but requires extreme async discipline.

Should we use Deel or hire offshore developers as direct contractors?

Use Deel if you have 5-plus offshore contractors or you need tax-compliance audit trail (SOC 2, enterprise sales). Use Wise plus a simple contractor agreement if you have 1 to 3 contractors and you are comfortable handling 1099-equivalent paperwork yourself. Use Cadence if you do not want to set up either and just want a vetted engineer with payment handled.

How do we handle code review when our offshore engineer is asleep during our workday?

Follow-the-sun: the engineer opens PRs at their end of day, the home team reviews during their workday, the engineer wakes up to actionable comments and iterates immediately. Enforce an 8-hour reviewer SLA and require CI to pass before review is requested. Median PR turnaround on this model is 24 to 36 hours from open to merge.

Is offshore cheaper than hiring locally?

Yes, typically 40 to 70 percent cheaper fully loaded. A Mid engineer at $1,000/week on Cadence (a typical offshore rate) is $52,000/year, versus $180,000 to $240,000 fully loaded for a comparable San Francisco hire. The savings only materialize if your onboarding works; a failed offshore hire is more expensive than a slow local hire. Our deeper analysis of remote engineering team cost savings breaks down the math.