** Gina Tsarin Software na Warkar da Kai tare da Multi-Agent AI Architectures da Orchestration Tools**

** Gabatarwa ***

A cikin yanayin yanayin software mai sauri da canzawa koyaushe, gina ƙaƙƙarfan tsarin daidaitawa wanda zai iya tsira daga samarwa shine babban fifiko ga injiniyoyin software da ƙungiyoyin DevOps. Hanyoyi na al'ada don haɓaka software galibi suna mayar da hankali kan gina tsarin monolithic waɗanda ke da saurin lalacewa da raguwar lokaci. Duk da haka, tare da zuwan gine-ginen AI masu yawa na wakilai da kayan aikin kade-kade, yanzu yana yiwuwa a ƙirƙiri tsarin software na warkar da kai wanda zai iya dacewa da yanayin canza yanayi da murmurewa daga gazawa.

A cikin wannan labarin, za mu bincika manufar gine-ginen AI mai wakilai da yawa da kayan aikin ƙungiyar, da kuma yadda za a yi amfani da su don gina tsarin software na warkar da kai. Za mu kuma zurfafa cikin sabbin abubuwan da ke faruwa a AI, injiniyan software, da DevOps, da kuma bincika rawar kayan aiki kamar LangChain da DSPy wajen ƙirƙirar tsarin software masu ƙarfi da daidaitawa.

** Menene Multi-Agent AI Architectures?**

Multi-agent AI gine-ginen nau'in tsarin AI ne wanda ya ƙunshi wakilai masu cin gashin kansu da yawa waɗanda ke hulɗa da juna don cimma manufa ɗaya. Kowane wakili yana da nasa tsarin iyawa, ilimi, da manufofinsa, kuma suna aiki tare don cimma manufa mafi girma. Gine-ginen AI mai wakilai da yawa suna da amfani musamman a cikin hadaddun tsarin inda akwai masu ruwa da tsaki da yawa, maƙasudai masu karo da juna, da sakamako marasa tabbas.

A cikin injiniyan software, ana iya amfani da gine-ginen AI masu yawa don gina tsarin warkar da kai wanda zai iya ganowa da amsa gazawar a cikin ainihin lokaci. Misali, tsarin AI mai wakilai da yawa zai iya ƙunsar wakilai da yawa waɗanda ke lura da sassa daban-daban na tsarin, gano gazawa, da kuma haifar da ayyukan dawo da tsarin zuwa yanayin lafiya.

Mene ne Kayan Aikin Orchestration?

Kayan aikin ƙungiyar ƙira tsarin software ne waɗanda ke sarrafa turawa, ƙira, da sarrafa aikace-aikace da ayyuka. Suna samar da dandamali mai mahimmanci don sarrafa abubuwa da yawa, ayyuka, da aikace-aikace, kuma suna ba wa masu haɓaka damar ayyana hadaddun ayyukan aiki da na atomatik.

Kayan aikin kayyade suna da mahimmanci don gina tsarin software na warkar da kai, yayin da suke baiwa masu haɓakawa damar ayyana ayyukan dawo da aiki da na'urori masu sarrafa kansa waɗanda za a iya haifar da su don amsa gazawar. Misali, ana iya amfani da kayan aikin kaɗe-kaɗe don sarrafa aikin tura tsarin ajiya, ko kuma haifar da aikin dawo da aiki lokacin da ɓangaren tsarin ya gaza.

** LangChain da DSPy: Kayan aikin Orchestration don Multi-Agent AI Architectures**

LangChain da DSPy sune shahararrun kayan aikin kaɗe-kaɗe waɗanda aka ƙera don yin aiki tare da gine-ginen AI masu yawa. LangChain shine ɗakin karatu na Python wanda ke ba da API mai sauƙi da ƙwarewa don gina tsarin AI mai yawa, yayin da DSPy ɗakin karatu ne na Python wanda ke ba da babban matakin API don ƙayyadewa da aiwatar da ayyukan aiki.

LangChain da DSPy suna ba masu haɓakawa damar gina tsarin software na warkarwa da kansu waɗanda zasu iya ganowa da amsa gazawa a cikin ainihin lokaci. Suna ba da fasali da yawa, gami da:

• ** Gudanar da wakili ***: LangChain da DSPy suna ba da APIs don ƙirƙirar, sarrafawa, da hulɗa tare da wakilai a cikin tsarin AI mai yawa. ** Ma'anar gudanawar aiki ***: LangChain da DSPy suna ba masu haɓaka damar ayyana hadaddun ayyukan aiki da na'urori masu sarrafa kansu waɗanda za a iya haifar da su don amsa gazawa.
• ** Ayyukan farfadowa ***: LangChain da DSPy suna ba da APIs don haifar da ayyukan dawowa, kamar ƙaddamar da tsarin ajiya ko haifar da aikin dawowa lokacin da ɓangaren tsarin ya kasa.

** Gina Tsarin Warkar da Kai tare da LangChain da DSPy ***

Don nuna ikon LangChain da DSPy, bari mu gina tsarin warkarwa mai sauƙi wanda zai iya ganowa da amsa gazawar a cikin aikace-aikacen yanar gizo. Za mu yi amfani da LangChain don ƙirƙirar tsarin AI mai wakilai da yawa wanda ke kula da aikin aikace-aikacen kuma yana haifar da aikin farfadowa lokacin da aka gano gazawar.

Anan ga ƙaƙƙarfan toshe lambar da ke nuna yadda ake gina tsarin warkar da kai tare da LangChain da DSPy: `` Python shigo da langchain daga Wakilin shigo da langchain, Sarkar

wakili = wakili ( name="wakilin-warkar da kai", sarkar= sarkar( suna = sarkar-warkar da kai", matakai=[ {"name": "mai dubawa", "func": Monitor_performance}, {"name": "farfadowa", "func": recover_system}, ], ), )

Ƙayyade aiki don saka idanu akan ayyukan aikace-aikacen

def Monitor_performance(): # Kwaikwayi gano gazawa idan bazuwar.random() <0.5: dawo {"kasa": Gaskiya} wani: dawo {"kasa": arya}

Ƙayyade aiki don dawo da tsarin

def recover_system( gazawa): # Fara aikin dawo da aiki idan gazawar[" gazawa"]: buga ("Yana jawo aikin dawo da...") # Sanya tsarin ajiya ko kunna aikin dawo da aiki #...

Ƙirƙiri aikin DSPy

aikin aiki = DSPy. Aikin aiki ( suna = "warkar da kai-gudanar aiki", matakai=[ {"name": "mai dubawa", "func": Monitor_performance}, {"name": "farfadowa", "func": recover_system}, ], )

Gudanar da aikin

aikin aiki.execute()

Wannan toshe lambar yana nuna yadda ake ƙirƙirar wakili na LangChain wanda ke lura da ayyukan aikace-aikacen kuma yana haifar da aikin dawo da aiki lokacin da aka gano gazawa. Ana amfani da aikin DSPy don ayyana aikin dawo da aiki da aiwatar da shi don amsa gazawa.

** Manyan Dokoki 5-10 da Akafi Amfani da su/Snippets**

Anan akwai manyan umarni 5-10 da aka fi amfani da su da snippets don LangChain da DSPy:

1. `langchain.Agent()': Yana ƙirƙira sabon wakilin LangChain.
2. `langchain.Chain()`: Yana haifar da sabon sarkar LangChain.
3. `dspypy.Tsarin aiki()`: Yana ƙirƙira sabon aikin DSPy.
4. `agent.execute()`: Yana aiwatar da wakilin LangChain.
5. `workflow.execute()`: Yana aiwatar da aikin DSPy.
6. `agent.chain.steps.append({"name": "step", "func": func})': Yana ƙara sabon mataki zuwa sarkar LangChain.
7. `workflow.steps.append ({"name": "mataki", "func": func})': Yana ƙara sabon mataki zuwa aikin DSPy.
8. `agent.chain.steps.remove({"name": "step")': Yana cire mataki daga sarkar LangChain.
9. `workflow.steps.remove({"name": "step")': Yana cire mataki daga aikin DSPy.
10. `agent.chain.steps.insert(0, {"name": "step", "func": func})': Yana saka sabon mataki a farkon sarkar LangChain.

**Kallon gani**

Anan ga hangen nesa na LangChain da aikin DSPy ta amfani da Chart.js:
Json
{
  "type": "bar",
  "data": {
    "labels": ["Duba", "Maida"],
    "datasets": [
      {
        "label": "LangChain",
        "bayanai": [10, 20],
        "backgroundColor": "rgba(255, 99, 132, 0.2)",
        "BorderColor": "rgba(255, 99, 132, 1)",
        "BorderWidth": 1
      },
      {
        "label": "DSPy",
        "bayanai": [20, 30],
        "backgroundColor": "rgba(54, 162, 235, 0.2)",
        "BorderColor": "rgba(54, 162, 235, 1)",
        "BorderWidth": 1
      }
    ]
  },
  "zaɓi": {
    "take": {
      "nuna": gaskiya,
      "rubutu": "LangChain da DSPy Workflow"
    },
    "ma'auni": {
      "yAxes": [
        {
          "zama": {
            "beginAtZero": gaskiya
          }
        }
      ]
    }
  }
}

Wannan hangen nesa yana nuna aikin LangChain da DSPy, tare da matakin "Duba" a hagu da matakin "Maida" a dama. Aikin LangChain yana wakilta ta mashaya ja, yayin da aikin DSPy ke wakilta ta mashaya shuɗi.

Kammalawa

Gina tsarin software na warkarwa da kai tare da gine-ginen AI masu yawa da kayan aikin kaɗe-kaɗe hanya ce mai ƙarfi don ƙirƙirar ƙaƙƙarfan tsarin daidaitawa wanda zai iya tsira daga samarwa. LangChain da DSPy sune manyan kayan aikin kade-kade guda biyu waɗanda ke ba masu haɓaka damar gina tsarin warkarwa da kansu waɗanda zasu iya ganowa da amsa gazawa a cikin ainihin lokaci.

A cikin wannan labarin, mun bincika manufar gine-ginen gine-ginen AI da yawa da kayan aikin ƙungiyar, kuma mun nuna yadda za a gina tsarin warkar da kai tare da LangChain da DSPy. Mun kuma bayar da misalai masu amfani da dama da snippets na lamba don taimakawa masu haɓakawa su fara gina nasu tsarin warkar da kansu.

Ko kai injiniyan software ne, injiniyan DevOps, ko mai bincike na AI, gina tsarin software na warkar da kai tare da LangChain da DSPy hanya ce mai ƙarfi don ƙirƙirar ƙaƙƙarfan tsarin daidaitawa wanda zai iya tsira daga samarwa.