Kwimeko ye-inDrive Product Design, i-AI yintloko esebenzayo. Isetyenziselwa iingxowa ze-UX ezisetyenziswa kumazwe ezahlukeneyo, ukutshintshiselana iingxaki ze-routine kwi-Figma, kunye nokukwazi ukuvelisa ngokukhawuleza iividiyo ze-realistic ukusuka kwi-illustrations. Ngezantsi zihlanganisa iingoma ezintathu zokuqinisekileyo: indlela i-designer yenza izixazululo ze-AI, iingxaki ezinxulumene kunye neziphumo ezifanelekileyo. Ukucaciswa ngaphandle kwe-Intermediaries - Ukucinga Polina Gladkova Polina Gladkova Kwiintsuku ezidlulileyo, iingxowa kunye neengxaki e-Egypt kunye ne-Latin America ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye ziye zibe. Ukwandisa ukucaciswa, i-Polina iye yenza iingxowa ngokufanelekileyo ngokusebenzisa i-ChatGPT ye-voice. I-Designer ibhalisele i-Russian, i-driver ibhalisele i-Arabic, i-driver ibhalisele, kwaye i-ChatGPT ibhalisele. Kwiimeko, iye ndingathanda ngathi: 
 
 
 
 
 
 
 preparing the interview script in ChatGPT beforehand; 
 
 during the ride, enabling voice chat and setting a translator prompt with dialect clarification (e.g., “Egyptian Arabic”) when needed; 
 
 after the interview, asking ChatGPT for: a detailed review of the conversation, a summary table for a series (e.g., 10 interviews), recurring patterns/differences, and hypotheses; 
 
 
 simultaneously recording audio → transcribing it in another AI tool → feeding the text into ChatGPT for more accurate processing. Izinzuzo : 
 
 
 
 nganye, i-ChatGPT iye iye iye iye iye iye iye iye iye iye zibonisa i-Russian phrases ngaphandle kokuguqulwa; Iingxaki zeengxaki zeengxaki zeengxaki zeengxaki zeengxaki zeengxaki zeengxaki zeengxaki zeengxaki zeengxaki zeengxaki zeengxaki zeengxaki zeengxaki zeengxaki zeengxaki; Ngexesha elide yeengxaki ze-interviews, iingxaki ezidlulileyo. Yintoni kunceda: 
 
 
 
 Ukubonisa i-dialect kwi-prompt; Ukusebenza nge-text (i-audio → transcript → ChatGPT); Manual monitoring ngexesha iingxowa. Ukusabela: 
 
 
 
 ixabiso ixabiso malunga ne-3-5x ekuqaleni kwe-traditional interpreter scheme; i-"cleaner" i-experiment: ukunciphisa i-influence ye-third parties, ukuxhaswa kwe-one-on-one; ekusebenzeni kwimveliso yobugcisa – ukusetyenziswa okuzenzakalelayo kwe-ChatGPT kwi-translations kunye ne-textual feedback on logic/UX. Okuzenzakalelayo Lokalization and Routine in Figma - Ukucinga U-Sergey Goltsov U-Sergey Goltsov Sergey usebenzise iingxaki ezininzi ngokwakha i-Figma plugins. Umgangatho onomgangatho: ukuthatha indawo yokuba (kufakelwa kwi-Figma forums / i-chats okanye kwimeko yobugcisa), ukulungiselela isicelo esilandelayo, kwaye ukwakha i-plugin usebenzisa i-combination ye-ChatGPT + i-Figma Plugin API documentation. Ngokutsho kwakhe, i-ChatGPT ikhiqiza ukuya kwi-80% yeenkcukacha; kwiminyaka engaphezu kwe-manual review and refinement (i-HTML/CSS/JS, i-testing kwi-editor kunye ne-Figma). Plugins ezibonakalayo ngokubanzi by Sergey: 
 
 
 
 
 
  — 9,500+ users,   in Figma Weekly. Chat Builder featured 
 
  — ~3,600 users. ChartBG 
 
 
  — ~2,600 users. BorderMockup One internal case was solving a recurring pain point: the monotonous manual work of creating translation keys and linking them to layers in Figma. To address this, Sergei built the   plugin. It scans the entire file (including groups, Frames, and Auto Layout), finds all text layers, and converts them into text variables. The plugin automatically cleans variable names according to API requirements; if the same text is repeated, only one variable is created and all relevant layers are linked to it. Text to Strings Ukwakhiwa kweChat Ukucinga Ukucinga BorderMockup Umxholo we Strings Umzekelo we-internal ye-automation ye-traductions e-Figma. Kwixesha elide, ukutshintshiselana kwe-layouts yaba umsebenzi oluthile ye-routine: i-texts kufuneka ifakwe ngempumelelo, i-versions ezahlukileyo ze-screens ziye ziye ziye zithunywe, kwaye i-updates ziye ziye zithunywe ngexesha lokuphinda. Ukukhusela le monotony, uSergei yasungula inkqubo esekelwe Figma Variables kunye nePlugin Sheet to Variables. Iingxelo ziye ngokuthe ngetyala kwiifayile ze-variable, ama-translators zokusebenza kuphela kwi-Google Sheets, kwaye umncedisi uxhumane iifayile ezongezelelweyo nge-CSV. Emva kokuba iifayile zihlanganiswe kwiihlabathi ze-layout, ukuguqulwa kwilwimi kwi-Figma kunzima kuphela iifayile ezininzi - kwaye zonke iingxelo zithuba ngexesha elifanelekileyo. Yintoni ukuchitha: 
 
 
 
 ukuvelisa izibuyekezo ezininzi ezininzi (ngokuthi "ngaba zonke iingxaki zokuzonwabisa ngaphambi kokwenza"); i-manual validate code kunye ne-cross-check kunye ne-documentation: i-model ingathintela i-logic yokusebenza okanye ibonise i-API ezidlulileyo; I-Chat Builder yokuzonwabisa lithunyelwe kwi-community posts kunye ne-industrial digest. Ukusuka kwi-illustrations ukuya kwiifoto ze-realistic - Ukucinga Arthur Sitdikov U-Arthur Sitdikov Ukusetyenziswa kwe-foodtech ye-inDrive, izibonelelo ze-flat ziye zisetyenziswa ngexesha elide. I-Arthur yaba ufuna ukuvavanya i-hypothesis: iifoto ze-realistic zeemveliso zithembisa kakhulu ngenxa yokuba abantu bayazi ngexabiso olufikelelekayo. Iifowuni ze-conversion ziyafumaneka, kodwa isisombululo esizayo - ukuvela ngokukhawuleza iimveliso zeemveliso zeemveliso - iye yenzelwe. Indlela Yenziwe: 
 
 
 
 ukuthatha izibonelelo ezikhoyo (i-banana, i-bread, njl) kunye nokufaka kwabo ngqo kwi-layout; Ukubiza i-ChatGPT ukuba "ukudlala" isakhiwo kwi-stylist; ukufumana iimveliso ezisetyenziselwa ukusetyenziswa kunye neziphondo ezininzi kunye nokusetyenziswa kwimveliso kunye ne-promo. Sample prompt: "Ngaphandle kokuguqulwa kwisombululo, kwenziwe ngokwenene, njenge-magazine. Ngaphandle kwe-transparent background." Yintoni ezaziwa: 
 
 
 iziphumo zokuqala zilungele kakhulu; ngexesha elide, iziphumo zangaphambili zangaphambili - ngokuvamile iiyure ezimbini ngalinye ibonisa; nganye iimodeli "ukugqiba" malunga ne-transparency kunye ne-checkboard pattern - oku kufuneka ifumaneke. Ukusabela: 
 
 
 
 
 
 first acceptable variants in ~15 minutes; 
 
 a complete set of assets in two evenings instead of lengthy approvals with photoshoots or stock purchases; 
 
 
 the approach is already in use in product/promo; in parallel, 3D images were generated for interfaces, and outputs from photoshoots were mixed with AI generations. I-Key Takeaways evela kwi-Three Cases 
 
 
 
 Kwi-interviews, i-Voice-based ChatGPT ivumela ukuqhagamshelwano ngqo kunye neengxaki, ukunyusa i-analysis, kunye nokunciphisa i-interpreter influence (ukugcina ixesha esilinganiselwe kwi-3-5x). Ukusetyenziswa kwe-layout ye-localization yaba ngempumelelo ngenxa ye-creation ye-variable ye-automatic kunye ne-CSV translation cycle (ukuguqulwa kwe-language kwi-one-click). I-Visual Generation iye yenza iimveliso ezininzi ezininzi kwiintsuku ezimbini, kwaye iimvavanyo ezimbini ezimbini - ezininzi yokubonisa iimvavanyo ngokukhawuleza kubathengi kunye nokukhuthaza. Ukucinga Zifumaneka ngezindlela ezintathu zokusetyenziswa kwe-AI kwimveliso yobugcisa: njenge-translator kunye ne-secretary ye-interviews, njenge-co-author ye-plugin code, kwaye njenge-tool ye-visual fast and realistic. Kwiimeko ezininzi, abaculi babonisa iingcaciso zabathandakanya kunye nokufumana njani. Iingcaciso ezininzi zibonakalayo: isantya eshushu, ukunciphisa iinkqubo ezincinane, kunye needatha ezincinane kakhulu ekwenzeni izixazululo zokusebenza.

inDrive.Tech

Le audio iveliswe ngolwimi lokuqala lwebali!

Inde kakhulu; Ukufunda

Boost your HackerNoon story @ $159.99! 🚀

I-AI kwi-Product Design: Amaxesha ezintathu ezisebenzayo ezivela kwi-inDrive

About Author

IZIMVO

ZIJONGE IIMPAWU

ELI NQAKU LINIKEZELWE KU

Related Stories

Light-Mode

Classic

Newspaper

Dark-Mode

Neon Noir

Minty

HN StartUps