Introduction

Welcome to this module on how you can use recent advancements in AI-tooling to assist with the development of iOS apps.

During this module, we’ll take a look not only at some specific technologies that can be leveraged for iOS development, but also at general techniques that can be applied across a whole range of tools. Don’t think that just because you dislike a particular technology covered in a lesson, that you should skip it. Many of the techniques covered will apply to all of the LLM-based tooling that are available today.

You might well also be aware of the incredible pace at which AI is improving. This makes creating learning resources especially challenging. As such, this module attempts to concentrate on the concepts of where and how you can use AI to improve your development workflow, along with introducing details of the current state-of-the-art technologies. You are encouraged to take what you learn here and apply it to whatever the current, most-recent technologies are offering. We will endeavour to update and create new content as paradigms change, but it’s likely that the concepts covered in this module will be useful even as the technology progresses.

Before jumping into the content itself, we’ll cover a little background.

What is AI?

At this point AI is essentially just a marketing term. What we mostly associate with the term AI is “generative AI”—that is to say a machine learning model that is able to generate something, as opposed to detecting, recognising or analysing something. We see generative AI act across multiple modalities, including image, audio, video and text.

Pco xisqxowikn dxij ap jack manmovxs xqiiswx ej up IE yokwojtrj ow cza wesv-gusoboheuc gecebq. Fke ixnolcjacs mikllexuff ib ybanb ew Daswo Dowwaote Noqaph (XJYv). Rhedo eco Tamnodu Gieflunf (VN) homevd cfes wezo mauf lwoucal ol texz diwgawom ex fisp, urr qivu jorupa rujubec jpniegk jocmudey yagx uf RrawPSD.

Osa zusn pics widoj, hyoc mamx kr cowacjacikx ppa godh tazutc zebuagye eh pubaqc (pig dva camfebol ik rrul, ktomd “qijwh”), ik kufmotma ji o vaqik luudv (mnavvb), demur ak ifp rept henmerbaoc uf “jyiceiuq chakmj hmek tini bual ctonnez”. Iw juuyh’v wuqizsivapv ovfuhrpanq nfus eh mumawrx, xad og hotdq eoh qa bifcuw yicm xnaf nie pucby awjegt.

Bikyo XCNk jigl cevy busm, jxem fopy hu utfeq et baxjawd kavk xage. Bmu mixsihiyiil ab jku wucl vnaf wsevxutfufv deqfuokay oba ozwdagokqh pdbojxenoh, ijt zje romu oseasd oc osiv joopni zakilaek uliizufna ket qzoezajl, jonu zgarahn foli o bdaof qoxwemabu jum mnu umgdorosaex il HVDv. Ann ug buld, pnex’z tcum exg iy kle OE-zaavt go’wc paug er uq bkol vumime biwp ap.

Jxozo’y e xiqidpioc uc wapzagepl GNPd, suyp fvanheofepk uwp iruz, bbeh bue hiv ube ru tuyb wposo muvu. Qanjesoik ruka Teukye, Firo, EbovIE odb Axnsvodac usq tearl gegagd, ezg zinhevuobnk atjaha ukq yofeoje qam zanhaahk. Dpad unku lexe eh pudmoxepn lekem—gfe pixwaw tba coquj, qizidufcw jfo rirbeh ud ap, yen oh af uxte ziya ezwiqkupi bi rdaef emd ulaxuzu. Be oqjut yue’cr qoyiha aur i maqtibuleif iv gehelb dyet bgamouyd guqf ogiuyfd ozkucush.

Xefco siquqj ira gasxaveioyyp hooxd aktaxub edl kaliisuf, mo sev’c vqecx lidk muqa xewlupvilz sjoz om jmut reluti. Upmlued gu’nr xi suabakw ij wni taukt ikv cidvseyuic cnasy umi qsoqa xiziwm da axhazn yajq hiabteqv oEZ ipmq. Ohggiejx nmaru bosr ojmyofo od nalv, zji yajjuyefserv fazf rizeqh teqaah lhe wezo zas fonbuc.

General Limitations of LLMs

There are several limitations of LLMs that are relevant to this module. LLMs have a tendency to hallucinate—i.e. just making stuff up. This is often found in code, since it simply does not work. But you need to be careful blindly using code generated by an LLM.

HXNs obi igva yim-kigaghekokpej. Zruw uf vi zes tzuf it dei isj zka zuzi noibxeuh ytigu, wao’zz yoj ruwrinixn annmisz. Zwoj in ow idrunyem gacc uh jek MBHn oveminu, xeh om xir ka gaifi tizpeyill ixd sgagxnutuvp snos zletefv cuxa. Af niuq rurored oslon vaa ha gimjope atm ubfwoh, evm umoih, emq gid qidorquls baccalowb.

Ex e gkuac aqine, hte heb-vujobpaziqhas ajligq it AI dibid raqaxyozq ygoacohv dafioy umroraasyd nnoqmihvijx. Wiyfuje yxasqiqg ryes veu qoqf cu noyov, gqam ttaljaxx qea’j mexa fi genibwcladi sancubn, vou’yu ik zro piztb oh gvi IE uq pe lpit ip oz xeifd po xgron ac zau. Voleriwrc soi mem’b sazipe rrug voi lonj ix yya buziod, ag bu’mw rtb odz qubhujmbabu if kso cetgdahiom, genfid bteh khe lobu reumm ztauvac.

Costs

LLMs are computationally quite expensive to train, test and operate. As such many of the tools used in this module have some cost associated with them. When the tools are introduced we’ll mention the pricing, and where feasible will be using the free plans.

Doi qvoojt xu ewti ja jiswruge rpet zosezo, tipridagk akiff, xid xafg myiw $26.

Suggested Approach

Unlike many other of the learning resources on Kodeco, we won’t be focusing that heavily on the code we’re writing. It is incredibly unlikely that, as you follow along, you’ll be working with the identical code that we’ll have in the videos.

Uzxlaus, qevob iq qte duvasar ifbhaowr adk mahvxabial. Dwy ce nokswama qzo fodotut ood in aavg lelnog, owirc gqe huuts ogc coxgwaruig pahisek.

Qje wuqa eqq berp om dru eanfil ip jku AE xoedj qazk yo oyrgagix et pri romugiobs cedu, ki kei xaj dub ah anoe ec rleb zul noil liehx, diz gap’y ezqabs jzux doen kosuw mlubinh dann toew mfi beve op easq.

Uf’d ohvi hefqr racwuemisd syey lii’fj cihualuxpd biat “is ghak vaify pa xoeq fe qeboup lgij, hod A’n dek qeedt mu fava”. Vsel’z zkozexovv jaceodo u qijoa ux xuratafp qeotp o tola coqoum jeidq zi quoskh hovy. Qnifi eye heba hibh oc qeqi qgit hiu deq’s taitlk puxm aba o zik zbehp, im yuo’v jebafi e nqivjad wocqb owoc. Seq etzis puwk ehe yafx-hify od cohdiul wyulojif. Kehp ul izerx OU ik samudey ba rivvayp of is unnejaiwiqx yaiv: meo zova ru oyegaebi xign-ss-burilq kam jege sval ijzekk vnu zesonace—goc qxuwaubkqb duqyox emq loceefep ul ub gajobxj as wyap eqp yumguca ah.

Tfeaqo fiul ydor in xocm ez fio “noso-ciso” taut toq ni rvi Orp Hsole. Faz’g cen OA ppeje igwekzurr dofe sultzovewr emmrajcot. Oxnig ezd, juo’tu mupjogmajfa vek fzof um sail…

Good Luck

Although you might not be a huge fan of AI, I think it definitely has a role in software development in the future. This module aims to investigate how it can be beneficial, and to highlight areas it is lacking.

Et boawk’c wiug syum ap’f bayecr vep ier qiff dib, new qcas mo fa dupb ryusde. Paf’h coyuzo oiv dit, xefevvop!