Hamilton Ridley Consulting
Hamilton Ridley Consulting
Hamilton Ridley Consulting
Field Notes · Vol. 05
When to Call for Help
For New Builders · 2026
A short guide

Solo is brave. Partnered is sustainable.

Daniel Kemp··8 min read

AI tools made you 10× faster. They didn't make you 10× more experienced. The gap between "I shipped this" and "should I have shipped this?" is where partners earn their pay. The skill that decides whether your project survives its first hard year is recognizing when you've hit a wall — and scoping the engagement so the help actually lands.

!
If launch is days away

If your client launch is less than a week out and nobody else has read the code, that is the wall. The choice is whether you bring help in before launch — or after the first incident. The first one costs $2,000. The second one costs your reputation with that client.

Lane 01
Signs of the wall
Frustration vs. real blocker
Lane 02
Kind of help
Match shape to problem
Lane 03
Scope it tight
Specific briefs, real outcomes
Lane 04
Relationship
Solo, project, retainer
§ 01

The four lanes

Lane 01

Signs you've hit a wall

Self-awareness is the hard part. The signals that mean "this is bigger than me" rarely announce themselves; they look like ordinary frustration until they aren't.

What changes between optionsFrustration vs. real blocker. Some bugs you're going to grind through. Some are systemic and no amount of solo grinding fixes them. Spotting the difference is the senior-engineer skill.
Same bug 3+ days running
Context exhaustion. The agent has been around the loop and you have too.
Pre-launch panic
"Is this actually safe?" If you can't answer with evidence, you're hoping.
Cost explosion you can't trace
Almost always an architecture issue. Solo debugging won't find it.
"Shipping it anyway"
The most expensive sentence in software. Stop.
Lane 02

Kind of help

Help isn't one thing. The kind that fits depends on what's blocking you — point-in-time review, ongoing access, pair debugging, or strategic call.

What changes between optionsHow focused vs. broad the help needs to be. Mismatched help (broad consulting for a narrow bug, narrow audit for a strategic decision) wastes both of you.
Code or pre-launch audit
Point-in-time, written report. Best for shipping with confidence.
Pair-programming session
Synchronous, learning while solving. Best for one stuck bug.
Architecture review
Strategic, before you build. Best when the next decision is hard to undo.
Retainer / Slack access
"Is this dumb?" answered in hours, not days. Best for ongoing work.
Lane 03

Scope it tight

Engagements that succeed are scoped. Open-ended ones drift, disappoint, and convince both sides the relationship doesn't work.

What changes between optionsHow specific the deliverable is. Vague briefs invite vague output. Specific briefs let your helper come back with something concrete you can act on.
Defined deliverable
Written report, fixed bug, recommended architecture. Not "advice."
Time-boxed
2 hours, 1 day, 1 week. Forces sharp scoping on both sides.
Single concern
One repo, one feature, one decision. Don't bundle.
Written follow-up
Email recap. No "did we cover X?" a week later.
Lane 04

Build the relationship

One-shot help works. Ongoing help compounds — second engagements are 5× faster than first ones because the helper already has your codebase loaded.

What changes between optionsWhether you treat the helper as a vendor or a partner. Vendors do the task. Partners catch the next problem before it costs you.
Project-based
Per audit, per consult. Lowest commitment, highest variance.
Monthly retainer
Predictable. They get your codebase loaded. You get them on Slack.
Async-first
Slack and email beat scheduling. Scheduling kills momentum.
Refer-up revenue
Pass jobs you can't take on, earn revenue share. Compounds incentive alignment.
§ 02

The same three projects, asking for help

Project A

Contractor lead form, week before launch

The form works. You think. There's a real domain pointed at it now. It feels different than "my localhost" and you want one more set of eyes before going live.

Sanity check, pre-launch
01 · Signal
Pre-launch nerves
Small project, real money on it, you can't fully justify your nerves but they're there. Trust the nerves; they're cheap to act on.
02 · Help
Standard Code Review
One-shot scored review applying the HRC v2.0 10-category rubric. Branded report, public badge, 30-min walkthrough. Match: narrow problem, narrow help.
03 · Scope
$2,000 · 5 days
Single repo, single commit anchor, written report + findings.json + badge. Defined and finite, no drift.
04 · Relationship
Project-based
Done in a week. Maybe come back later for the next project. Low commitment, real value.
Project B

Multi-user CRM, six months in

Eight people log in daily, the codebase is too big to keep all in your head, and you wake up worrying about edge cases. The solo phase served you well — it's stopped working.

Sustainable partnership
01 · Signal
"Too big for solo"
Real revenue, real users, a feature backlog that outpaces what you can review carefully. You're not stuck — you're scaling beyond solo.
02 · Help
Partner Retainer
Slack access for "is this dumb?" questions. One Standard Code Review per quarter for the major releases. Match: ongoing work, ongoing access.
03 · Scope
$599/mo · 1 review/quarter
Predictable, async-first, includes one Standard Code Review per quarter plus perpetual badge freshness. Scoped per month, not per question.
04 · Relationship
Long-running
Second-month engagements run 5× faster — the partner already knows your codebase. Compounds in your favor.
Project C

Nightly automation, mystery failure

Logs say success. Output is wrong. You've spent two days alone with the code and you're starting to repeat yourself. Time to bring in fresh eyes.

One specific bug
01 · Signal
Day 3 of same bug
Context exhaustion — you've been around the loop and so has the agent. You're not going to find it on day 4 either.
02 · Help
Pair-programming session
Synchronous, time-boxed, fresh eyes on a specific problem. Match: narrow blocker, focused help.
03 · Scope
2 hours, written recap
Single bug, single repo. If it can't be solved in 2 hours, that itself is the finding. Time-boxed protects both sides.
04 · Relationship
Project-based now, retainer if it recurs
One bug = one consult. Pattern of bugs = upgrade to retainer. Let the actual workload pick the relationship shape.
Ready when you are

Two ways to bring HRC in.

Most partners start with a free Quick Score, upgrade to a Standard Code Review for their first paid project, then move to the Retainer once they're shipping quarterly. All are month-to-month — the only commitment is to the project you're actually launching.

Standard Code Review
$2,000 / project

Full HRC v2.0 10-category review. Scored, branded HTML deliverable + findings.json + public badge. 30-min walkthrough. 5 business days.

Buy a Standard →
Partner Retainer
$599 / month

Slack access. One Standard Code Review per quarter included. Badge stays fresh perpetually. Referral revenue on jobs too big for you. Cancel anytime.

Start retainer →

New here? Read the full Partner Program overview — including the free tier (Field Notes + automated scan) you can start with at no cost.

SharePostLinkedIn
Field Notes Subscribers

Future series in your inbox the day they ship.

Vol. 01–05 covered the lifecycle of a vibe-coded project. The next series goes deeper on specific stacks, AI-agent patterns, and growth playbooks. One email per release. Unsubscribe in one click.

Hamilton Ridley Consulting · 2026
daniel.kemp@hamiltonridley.com