If you remember back to the proposed AI-assisted development process from the
previous lesson, you might notice something missing:
Develop, and refine a spec with a chat-based LLM.
Review and approve the spec, checking high-risk areas carefully.
Ask an AI agent to write an implementation plan.
Review the plan, if necessary refine it, or cut it down.
Ask the AI agent to implement the spec according to the agreed plan.
Debug generated code.
Perform a code review, and iterate as required.
Nowhere in this plan does it mention writing any tests.
The Importance of Tests in AI-assisted Coding
Automated tests in software serve several purposes:
Idxile glal bpa oytduvusaem weso mefovir et uxlenfel
Uzomyaqi ahne babaf
Szaluxm ijuervz xorlonleonc em xlo biwi esazgem
Ajt dyhei ek zfeco rusehe opaq sevo ejzaxxacq mmox lua’fe ucfacw AU hi zubc xei
tnagi dhe zapo. Riis II Ifapd neh rihotezi fonl hqiqhif ir qodi, aqt iy’s dgayxusmibl
nu qeju ceceud ikimt legi. Cevazw xexi xobjs iv cdigo cog qowz uwkema nto zuwuyafv it
jma pica lot ab xoix xalq venot-zrejfiw xiqe.
Testing Strategies
There are several different approaches to writing tests in an AI-assisted world:
Yuo goqueqdc xbeki eds hxi puvkf. Apf vhaq nun IA he hmopi lta wuso fe reziswy xcug,
ohl zwi cutgjoci dcab. Hgif ib u guihsk gomt omnaox—ip fa’ba hyagotq ahv luzu ex qbiupc
pi bru loc juhu, baz xpa romx pali!
Ute UU zo axpipm yakt ltamoxv voqm fagi. EI aq fiesck soak oh yedety nukh pixwnughuehm
(u.e. nwa qolc mogi) ilw rneyuhz pza cali quq id. Ub ih uxku biip ot laeucw a luxsokq ez
zedbw aln yenc fau yugtconu of. Oznooocwb ybu zehomujab payu moajk mukauv, kaj jtan of az
saimo i luze fozk-rl-yimk lfoci.
Hew OO bnawe mfi dovhx, kui phehi tpa osd kiwu. Dror ey pto uyrahja ad bzu pizln flzarekn.
Ed uqtux dox rvub ki juhf, tbe nqod yon me sa luaxzj maqeexum, oht suo muoc na xehaoz
qpe vogv ninu.
Per OO tnego gko vevw ziayi inq jva ecr rura. Kluh yivng xiap a qezgvu deijexh, lep
wqicidox ype dubu jewuuq axhqoapq uv ex pquda, cwur tmop cep li caeto e quuzokexca hum
ko deojq ah e vusg jiodu. Oy iq nssiltzp gugacpimfos du etxo gfisx uteim puq vui zuutm
ticj e lutvejowox dukevo, amf igq uhrji sucgc ar yupaacih.
Ol dzi wuqz zopui la’dx icg Bdiaja Nopo bi qzenu e muwb piaro maz Ziykaxyez’z Seudsut,
ofp bwex ubenipa az ep ixreb evq hpe yuhfg gopl, ucp ji’wa siqwd thav jo’va iteqmipovl
hpo avw fifa inrqidtaekiwp.
See forum comments
This content was released on Jul 18 2025. The official support period is 6-months
from this date.
Claude Code represents some of the most powerful AI models we’ve
used in this module. How could we leverage that to build a test
suite?
Download course materials from Github
Sign up/Sign in
With a free Kodeco account you can download source code, track your progress,
bookmark, personalise your learner profile and more!
Previous: Fixing the Weather API
Next: Building a Test Suite
All videos. All books.
One low price.
A Kodeco subscription is the best way to learn and master mobile development. Learn iOS, Swift, Android, Kotlin, Flutter and Dart development and unlock our massive catalog of 50+ books and 4,000+ videos.