I-Poolside AI Code Generation
I-Poolside iyimodeli yesisekelo sesisekelo se-AI esixhaswe kahle esikhethekile sokuthuthukiswa kwesoftware.
Uhlolojikelele
Its big bet is that training on real software-engineering feedback, not just scraped code, will produce models that out-code general-purpose LLMs.
I-Deep Dive
Yasungulwa ngo-2023 ngu-Jason Warner (owayekade eyi-GitHub CTO) kanye no-Eiso Kant, i-Poolside izimisele ukwakha amamodeli asemngceleni ahloselwe ngokukhethekile ikhodi esikhundleni sezingxoxo. Umbono wayo wesiginesha uthi Ukuqiniswa Ukufunda kusuka ku-Code Execution Feedback (RLCEF): esikhundleni sokubikezela ithokheni elandelayo kuphela, imodeli ibhala ikhodi, iyiqhube ngokumelene nokuhlolwa nabahlanganisi, futhi ifunde ukuthi isebenze ngempela yini. I-Poolside inyuse cishe amaRandi ayizigidi ezingama-626 ochungechungeni lwe-B lwango-2024 ngenani lamaRandi ayizigidi eziyizinkulungwane ezintathu, nabaxhasi abahlanganisa iBain Capital Ventures futhi kamuva noNvidia. Inkampani ithengisela amabhizinisi afuna amamodeli ekhodi atshalwe endaweni yawo, egcizelela ubumfihlo, ukusingathwa kwasendaweni noma okuyimfihlo, kanye nabasizi abaqondiswe kumakhosombe angaphakathi ekhasimende kune-API yomphakathi eyabiwe.
I-Technical Insight
I-RLCEF iphatha i-compiler ne-test suite njengesignali yomvuzo ozenzakalelayo. Imodeli ikhiqiza izixazululo zekhandidethi, izenze, futhi ukufunda kokuqinisa kuphusha izisindo emiphumeleni ehlanganisa futhi ephumelele ukuhlolwa. Ngenxa yokuthi ukulunga kungahlolwa ngokohlelo, i-Poolside ingakwazi ukukhiqiza impendulo yokuqeqeshwa yokwenziwa engenamkhawulo ngaphandle kwamalebula abantu, iluphu ehlasimulisayo lapho ukuqeqeshwa okuhlanzekile kwethokheni elandelayo kumakhosombe ekhodi emile angeke kukwazi ukukunikeza ngokwawo.
I-Strategic Impact
Isu lomthengisi
Imephu yemigwaqo yabathengisi ithonya ukuthi yiziphi izici iqembu lakho elingazakha ngokulandelayo.
Izindleko kanye nesabelomali
Imigomo yezohwebo nezinketho zokuthunyelwa zithinta izindleko zesikhathi eside nobungozi.
Ingozi nokuphepha
Izinxephezelo zenkampani zibumba okuzenzakalelayo komkhiqizo, ukuma kokuphepha, nokuvuleleka.
Ikusasa Le-Poolside AI Code Generation
I-Poolside igijima izimbangi ezifana ne-OpenAI, Anthropic, kanye ne-Cursor ukuze ube ngumnikazi wokukhiqiza ikhodi yebhizinisi. Lindela amakhono e-ejenti ajulile (ukuhlelwa kwamafayela amaningi, ukuqedwa komsebenzi ozimele), ukuthunyelwa okuqinile endaweni ezimbonini ezilawulwayo, kanye nokukala kwekhompuyutha okusekelwe yi-Nvidia. Umbuzo obalulekile ukuthi ingabe imodeli yesisekelo sekhodi kuphela ingahlala iphambili kunamamodeli avamile ahlala ethuthuka ezinhlelweni, kanye nokuthi amabhizinisi akhokha iprimiyamu yini yobumfihlo nokwenza ngokwezifiso.
Ukuqaliswa Komhlaba Wangempela
Ukuthumela umsizi wekhodi yangasese ngaphakathi kwengqalasizinda yebhange ukuze ikhodi yomthombo wobunikazi ingashiyi i-firewall.
Ukukhiqiza nokuqinisekisa ngokuzenzakalela ukuhlola kweyunithi ngokukusebenzisa ku-sandbox ngaphambi kokukuphakamisa konjiniyela.
Ukusiza ibhizinisi ukuba lenze isisekelo sekhodi yefa sibe sesimanjemanje ngeziphakamiso eziyimodeli ezishunselwe kulabhulali yangaphakathi yaleyo nkampani.
Ukunikeza usizo lokuqedela ngokuzenzakalela nokusekelwe engxoxweni olulungiswe kahle kumakhosombe athile ekhasimende nezimiso zokubhala amakhodi.
Izingozi & Guardrails
Izimemezelo zokwethula zingase zeqe ukuzinza ekugelezeni komsebenzi wangempela wokukhiqiza.
Izintengo ze-API noma izinguquko zenqubomgomo zingaphula ukucabanga ngobusuku obubodwa.
Ukuncika komthengisi oyedwa kukhulisa izindleko zokukhiya nokufuduka.
Ukuqalisa Umhlahlandlela
Linganisa abahlinzeki usebenzisa eyakho imisebenzi namasethi edatha.
Buyekeza ubumfihlo, ukuphepha, nemibandela yomthetho ngaphambi kokuhlanganiswa.
Gcina uhlelo lokubuyela emuva kuwo wonke amamodeli noma abathengisi.
Gada amanothi okukhululwa ukuze izinguquko zemephu yomgwaqo zingamangazi amaqembu.
Qhubeka Uhlole
Free newsletter
Get the daily AI briefing
Three verified AI stories every weekday morning, written in plain English. Free forever, no ads.
One email each weekday. Unsubscribe in one click. We never sell or share your address.
Test yourself
Take the Poolside AI Code Generation quiz
Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.
Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation
Umhlahlandlela olandelayo
Ukuqedwa Kwekhodi Ye-Tabnine
Imibuzo evame ukubuzwa
What is Poolside AI Code Generation?
I-Poolside iyimodeli yesisekelo sesisekelo se-AI esixhaswe kahle esikhethekile sokuthuthukiswa kwesoftware. Ukubheja kwakho okukhulu ukuthi ukuqeqeshwa ngempendulo yangempela ye-software-engineering, hhayi nje ikhodi ekhishwe, kuzokhiqiza amamodeli akhipha ama-LLM enhloso evamile.
Iyiphi indlela yokuqeqesha ewumongo i-Poolside eyaziwa ngayo?
Indlela yesiginesha ye-Poolside i-RLCEF, lapho imodeli isebenzisa khona ikhodi yayo ngokumelene nezivivinyo nabahlanganisi futhi ifunda ukuthi ingabe yenziwe ngendlela efanele yini.
Omunye wabasunguli bePoolside phambilini ubebambe yiphi indima?
UJason Warner, umsunguli we-Poolside, phambilini wayeyiChief Technology Officer ye-GitHub.
I-Poolside igcizelela ini kumakhasimende ebhizinisi?
I-Poolside iqondise amabhizinisi anokusingatha okuyimfihlo kanye nasendaweni ukuze ikhodi yomthombo wobunikazi ingashiyi indawo yekhasimende.
Iyiphi inkampani enkulu yama-chip eyaba ngumtshali-zimali/umsekeli ophawulekayo we-Poolside?
U-Nvidia waba umsekeli we-Poolside, esekela izidingo zayo ezinkulu zekhompiyutha zokuqeqesha amamodeli wekhodi yemingcele.
Kungani impendulo yokusebenzisa ikhodi ibaluleke kangaka ekuqeqesheni imodeli yekhodi?
Ngenxa yokuthi abahlanganisi nokuhlola kuqinisekisa ukulunga ngokohlelo, imodeli ingakhiqiza inani elikhulu lempendulo yokuqeqeshwa okuzikhiqizele yona ngaphandle kwamalebula omuntu abizayo.