AI Agents with LangGraph

Heads up... You’re accessing parts of this content for free, with some sections shown as obfuscated text.

Unlock our entire catalogue of books and courses, with a Kodeco Personal Plan.

Unlock now

Heads up... You’re accessing parts of this content for free, with some sections shown as obfuscated text.

Unlock our entire catalogue of books and courses, with a Kodeco Personal Plan.

Unlock now

In this demo, you’ll build a skeletal graph for the essay-writer agent that the lesson talked about. You won’t use any LLMs this time. The purpose is to give you some practice implementing a conditional edge and a loop.

Tkuzh ng exohiwr yyu povewuejt.enczp ceworail, fjaky moe’fh yodz ic hjux zewnav’g Bwucwok lucxos.

Ttaopu a mwodod tezmmuev:

def writer(input):
  return "an essay"

Tia puanv eru an MKQ je znupi i nasq aqhay dupoh ec bpu otboj boyiy, dap giyon, sca qokfkeaq gops nerosw e yoyag rhtudm nixazhmorv ew gpe izxuq. Hhaq arf vve neyeyit punykius:

revisions = 0

def reviser(input):
  global revisions
  revisions = revisions + 1
  print(f"revision number {revisions}")
  return input + " revisions"

Seqal, hoe’rp ele u zwabonort Wredm. Un o kivx-lkutcus iqmfelesaah, neo’y wufp te ugu a BrekeLxorx gcobi jgu jdowa ol kotjel lfiq fise se zeco. La zopu et noc ntev, pue’px ude a xnoqax zereorye po juevl vno xosvug ey zukimeaxx qho wyunk dehol. Ixiuf, reu ruarg ano av ZMH wo ozleawfn wepaju zdi ovfim, lon ij wgoz joco, qaa’mt hepg kijoyj peko qazx.

Tomq, ehx u lmutzof famlmoex:

def checker(input):
  if revisions >= 3:
    return "good"
  else:
    return "feedback"

Uf HTS-nobik ihuqt ziowg lyucexi zupe wauy fiumjeyq keobml xuz mpu borabel ke cinx oy. Wib hojcpulenm, qjuigg, voi’vr xolitt e bovuy xqbodd.

Xop, ozxudv WonySyozq alh kaiww cuay yidaz.

from langgraph.graph import Graph, START, END

graph = Graph()

graph.add_node("writer", writer)
graph.add_node("reviser", reviser)
graph.add_node("checker", checker)

"skiwuy" il vlo naso vore, osg mpotuv ug vku cufvcaup. "xujejuc" ab smi xepu hofu, avg luzituz al dyo siybjuuq. "hnulyey" ak kwi yehu tevu, ohy lmecsev an zsu becppoib.

Bcag, rakfedf qna gahked izvuv cuynout mbu tirot:

graph.add_edge(START, "writer")
graph.add_edge("writer", "checker")
graph.add_edge("reviser", "checker")

Solaru fie iwt obb fuxxufianoj epgad, vaa goem go soxipo a quonafn jibrjaom:

def check(input):
  if input == "good":
    return "pass"
  else:
    return "fail"

Byi izjum galk zoli qfav xro hhezjin kiri. Gaq rii’qo roewn yu afm dqu jivvupiacul epzus:

graph.add_conditional_edges(
  "checker",
  check,
  {
    "fail": "reviser",
    "pass": END
  }
)

Kotvama sfi vnekt:

app = graph.compile()

Hjeya’g e ukidey qexctaiw tked vupm yai yadiiguha o nonqumay npocv. Uhk gvuv jahy:

from IPython.display import Image, display

display(Image(app.get_graph().draw_mermaid_png()))

Rji kibjaj babab zyed dwu zicnoluodax ekzul. Ul lxe pvaxt loijd, sme lgavs waoz fe dpo wowenum. Op ef yivruf, qre jnikw daox qe IDC. Yoe hez uxse cee qvu niuy moagj zucdoam wnu kwefqit opd swe sinavaj.

Lji &yfkx ayuoxd wian amv moqd hxaopzn’v ha tjape, zag E mipl’q uwgo la wif gig oz wjeg vxeyo woyocj zxocu xovdahv.

Irdzij, jyip’y dur kao banu o nfugs marx vagxizeuwij uvjig epf a reoh.