Ƙirƙirar Haɓaka Ƙarfafa Software na AI tare da LangChain da StackGen | B4N1

Ƙirƙirar Haɓaka Ƙarfafa Software na AI tare da LangChain da StackGen

3 months ago · b4n1 Tiempo de lectura: 6 min

Gabatarwa

Yanayin haɓaka software yana haɓaka cikin sauri, tare da AI da aiki da kai suna ƙara yin fice a cikin tsarin DevOps. Mabuɗin fasaha guda biyu waɗanda ke jagorantar wannan canji sune LangChain da StackGen. LangChain tsari ne na bude-bude don gina ma'aikatan da ke da ikon AI, yayin da StackGen kayan aiki ne don daidaita tsarin kayan aiki. A cikin wannan labarin, za mu bincika yadda za a iya amfani da waɗannan fasahohin guda biyu tare don ƙirƙirar bututun sarrafa kayan aiki na DevOps na AI, wanda ke ba da ƙarfin GPT-5.5 don nazarin lambar fasaha da shawarwari.

Kalubalen DevOps Automation

DevOps aiki da kai aiki ne mai rikitarwa wanda ke buƙatar haɗa kayan aiki da fasaha da yawa don daidaita ci gaban software, gwaji, da turawa. Hanyoyi na al'ada sau da yawa sun dogara da hanyoyin hannu, wanda zai haifar da kurakurai, jinkiri, da rashin aiki. Yunƙurin AI da aiki da kai yana ba da mafita mai ban sha'awa, amma yana buƙatar zurfin fahimtar fasahar da ke ƙasa da haɗin kai.

LangChain: Wakilin Orchestration na AI-Powered DevOps

LangChain wani tsari ne na bude-bude don gina ma'aikatan AI-powered wanda zai iya yin ayyuka da yawa, daga sarrafa bayanai zuwa sake duba lambar. Ƙarfin ƙungiyar ƙungiyar sa na wakili yana ba masu haɓaka damar ƙirƙirar hadaddun ayyukan aiki waɗanda ke haɗa samfuran AI da kayan aikin da yawa. An ƙera gine-ginen LangChain don ya zama na zamani sosai kuma mai iyawa, yana mai da shi kyakkyawan zaɓi don gina bututun sarrafa kansa na DevOps na al'ada.

LangChain Architecture

Gine-ginen LangChain ya ƙunshi manyan sassa uku:

  1. ** Agents ***: Waɗannan su ne ainihin abubuwan da ke cikin tsarin LangChain, wanda ke da alhakin yin ayyuka na musamman kamar sarrafa bayanai, sake duba lambar, da gwaji.
  2. Tsarin aiki: Waɗannan su ne jerin ayyukan da wakilai ke aiwatarwa don cimma wata manufa ta musamman, kamar ginawa da ƙaddamar da aikace-aikacen software.
  3. **Orchestrator ***: Wannan shine bangaren da ke tafiyar da ayyukan aiki, tabbatar da cewa an aiwatar da ayyuka cikin tsari mai kyau da kuma cewa duk wani kurakurai ana sarrafa su yadda ya kamata.

Lambar Misali: Ƙirƙirar Wakilin LangChain

Ga misali mai sauƙi na ƙirƙirar wakili na LangChain wanda ke yin bitar lambar ta amfani da GPT-5.5: `` Python shigo da langchain daga langchain.chains shigo da gpt5_5

Ƙirƙiri wakili na GPT-5.5

wakili = gpt5_5.GPT5_5Agent()

Ƙayyade aikin bita na code

aikin aiki = langchain. Gudun aiki ( suna = "code_review", ayyuka =[ { "type": "text", "input": "Don Allah a sake duba lambar mai zuwa", "output": "code_review" }, { "type": "text", "input": "code_review", "output": "reviewed_code" } ] )

Ƙirƙiri wakilin LangChain wanda ke yin aikin bitar lambar

wakili = langchain.Agent (gudanar aiki = gudanawar aiki, wakili = wakili)

Gudanar da aikin sake duba code

sakamako = agent.run(shigarwar = "Don Allah a duba lambar mai zuwa") buga(result.output)

Wannan lambar tana ƙirƙirar wakili na LangChain wanda ke yin bitar lambar ta amfani da GPT-5.5. An bayyana wakili a matsayin tsarin aiki wanda ya ƙunshi ayyuka biyu: ɗaya wanda ke sa mai amfani ya ba da lambar don dubawa, da kuma wani wanda ke duba lambar ta amfani da GPT-5.5.

## StackGen: Kanfigareshan Kayan Aiki don DevOps Automation

StackGen kayan aiki ne don daidaita tsarin samar da ababen more rayuwa, yana mai da shi kyakkyawan zaɓi don bututun sarrafa kansa na DevOps. An tsara tsarin gine-ginensa don ya zama na zamani sosai kuma za a iya cirewa, yana barin masu haɓakawa su ƙirƙira saitunan abubuwan more rayuwa na al'ada don aikace-aikacen su.

### StackGen Architecture

Tsarin gine-ginen StackGen ya ƙunshi manyan sassa uku:

1. **Masu Ba da Agaji**: Waɗannan su ne abubuwan da ke sarrafa abubuwan more rayuwa, kamar AWS ko GCP.
2. **Configuration**: Waɗannan su ne ma'anoni na abubuwan more rayuwa, kamar na'urori masu mahimmanci ko bayanan bayanai.
3. ** Orchestrator ***: Wannan shine bangaren da ke tafiyar da gyare-gyare, yana tabbatar da cewa an ƙirƙira kayan aikin da kuma sabunta su daidai.

### Lambobin Misali: Ƙirƙirar Kanfigareshan StackGen

Ga misali mai sauƙi na ƙirƙirar tsarin StackGen don injin kama-da-wane:
``tsatsa
amfani stackgen :: config;

// Ƙayyade ƙayyadaddun tsarin injin kama-da-wane
bari vm_config = saitin :: VMConfig {
    suna: "my_vm".to_string(),
    mai bayarwa: "aws".to_string(),
    yanki: "us-west-2".to_string(),
    Misali_type: "t2.micro".to_string(),
    image_id: "ami-abc123".to_string(),
};

// Ƙirƙiri daidaitawar StackGen
bari config = saitin :: Config {
    suna: "my_config".to_string(),
    mai bayarwa: "aws".to_string(),
    yanki: "us-west-2".to_string(),
    albarkatun: vec![vm_config],
};

// Ajiye sanyi zuwa fayil
config.save("my_config.yaml");

Wannan lambar tana ƙirƙirar saitin StackGen don injin kama-da-wane ta amfani da akwatin 'stackgen'. An bayyana tsarin daidaitawa azaman tsarin `VMConfig', wanda ya haɗa da kaddarori kamar suna, mai bayarwa, yanki, nau'in misali, da ID na hoto.

Haɗin LangChain da StackGen don AI-Driven DevOps Automation

Don ƙirƙirar bututun sarrafa kansa na DevOps na AI, muna buƙatar haɗa LangChain da StackGen. Wannan ya haɗa da ƙirƙirar aikin aiki wanda ke amfani da wakilai na LangChain don yin ayyuka kamar nazarin lamba da gwaji, da StackGen don sarrafa albarkatun ababen more rayuwa.

Lambar Misali: Haɗa LangChain da StackGen

Ga misali mai sauƙi na haɗa LangChain da StackGen don ƙirƙirar bututun sarrafa AI na DevOps: `` Python shigo da langchain daga langchain.chains shigo da gpt5_5 shigo da stackgen

Ƙirƙiri wakili na GPT-5.5

wakili = gpt5_5.GPT5_5Agent()

Ƙayyade aikin bita na code

aikin aiki = langchain. Gudun aiki ( suna = "code_review", ayyuka =[ { "type": "text", "input": "Don Allah a sake duba lambar mai zuwa", "output": "code_review" }, { "type": "text", "input": "code_review", "output": "reviewed_code" } ] )

Ƙirƙiri wakilin LangChain wanda ke yin aikin bitar lambar

wakili = langchain.Agent (gudanar aiki = gudanawar aiki, wakili = wakili)

Ƙayyade tsarin StackGen

vm_config = stackgen.config.VMConfig { suna: "my_vm".to_string(), mai bayarwa: "aws".to_string(), yanki: "us-west-2".to_string(), Misali_type: "t2.micro".to_string(), image_id: "ami-abc123".to_string(), }

Ƙirƙiri saitin StackGen

config = stackgen.config.Config { suna: "my_config".to_string(), mai bayarwa: "aws".to_string(), yanki: "us-west-2".to_string(), albarkatun: vec![vm_config], }

Ajiye sanyi zuwa fayil

config.save("my_config.yaml");

Gudanar da aikin sake duba code

sakamako = agent.run(shigarwar = "Don Allah a duba lambar mai zuwa") buga(result.output)

Yi amfani da tsarin StackGen don ƙirƙirar albarkatun ababen more rayuwa

stackgen.create_resources(tsarin)

Wannan lambar ta haɗe LangChain da StackGen don ƙirƙirar bututun sarrafa kayan aiki na DevOps da AI ke motsawa. Bututun yana amfani da wakilin LangChain don yin bita na lamba ta amfani da GPT-5.5, da tsarin StackGen don sarrafa albarkatun ababen more rayuwa.

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

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

1. `langchain.Agent (aiki = gudanawar aiki, wakili = wakili)': Ƙirƙirar wakili na LangChain wanda ke yin aikin aiki ta amfani da takamaiman wakili.
2. `stackgen.config.VMConfig(...)`: Yana bayyana tsarin na'ura ta amfani da StackGen.
3. `langchain. Aiki (...)': Yana bayyana tafiyar aiki ta amfani da LangChain.
4. `stackgen.create_resources(config)': Yana ƙirƙira albarkatun ababen more rayuwa ta amfani da tsarin StackGen.
5. `langchain.Agent.run(shigarwa = "..."): Yana gudanar da wakili na LangChain tare da takamaiman shigarwa.
6. `stackgen.config.save("..."): Yana adana tsarin StackGen zuwa fayil.
7. `langchain.Workflow.tasks(...)': Yana bayyana ayyuka a cikin aikin LangChain.
8. `stackgen.config.resources(...)`: Yana bayyana albarkatu a cikin tsarin StackGen.
9. `langchain.Agent.agent(...)': Samun wakilin da ke da alaƙa da wakilin LangChain.
10. `stackgen.config.provider(...)': Yana samun mai bada alaƙa da tsarin StackGen.

## Kallon gani

Anan ga hangen nesa na LangChain da StackGen ta amfani da Chart.js:
Json
{
  "type": "bar",
  "data": {
    "labels": ["LangChain", "StackGen"],
    "datasets": [
      {
        "label": "Components",
        "bayanai": [3, 3],
        "Bayan launi": ["#4CAF50", "#FF9800"]
      },
      {
        "label": "Ayyuka",
        "bayanai": [5, 5],
        "Bayan launi": ["#2196F3", "#9C27B0"]
      }
    ]
  },
  "zaɓi": {
    "take": {
      "nuna": gaskiya,
      "rubutu": "LangChain da StackGen Architecture"
    },
    "tatsuniya": {
      "nuni": gaskiya
    }
  }
}

Wannan hangen nesa yana nuna adadin abubuwan haɗin gwiwa da ayyuka a cikin gine-ginen LangChain da StackGen.

Kammalawa

A cikin wannan labarin, mun bincika yadda LangChain da StackGen za a iya amfani da su tare don ƙirƙirar bututun sarrafa kayan aiki na DevOps na AI. Mun ga yadda za a iya amfani da ikon ƙungiyar ƙungiyar wakilai na LangChain don yin ayyuka kamar nazarin lamba da gwaji, da kuma yadda za a iya amfani da iyawar daidaitawar kayan aikin StackGen don sarrafa albarkatun ababen more rayuwa. Mun kuma ga yadda za a iya haɗa waɗannan fasahohi guda biyu don ƙirƙirar bututun sarrafa kansa na DevOps mara sumul.

¿Tienes preguntas?

Hablemos

Contáctame en LinkedIn