Aller au contenu principal

Optimize

Category: AI & ML
Node id: optimize

AI-powered optimization analysis with roadmap and quick-win prioritization

What it does​

  • Multi-category optimization (performance, cost, scalability, security)
  • Effort, risk, and impact assessment per recommendation
  • Quick-win prioritization with implementation roadmaps
  • Tradeoff and dependency analysis
  • Budget and timeline-aware recommendations
  • Overall optimization score calculation

Use cases​

  • Optimizing cloud infrastructure costs and resource usage
  • Improving pipeline execution performance
  • Security posture assessment and hardening recommendations
  • Scalability analysis for growing healthcare workloads

Runtime​

  • Performance tier: variable

Input ports​

NameTypeRequiredDescription
targetanyYesCode, config, workflow, or system to optimize
contextanyNoAdditional context
current_metricsanyNoCurrent performance/cost metrics
constraintsanyNoOptimization constraints
goalsanyNoSpecific optimization goals
budgetanyNoBudget constraints
timelineanyNoImplementation timeline
stakeholdersanyNoStakeholder requirements

Output ports​

NameTypeRequiredDescription
resultjsonNoMain result containing optimizations, summary, roadmap, and quick_wins
metadatajsonNoExecution metadata including model, tokens, overall_score, and counts
successbooleanNoWhether the operation succeeded

Settings​

Basic​

SettingNameTypeRequiredDefaultAllowed valuesDescription
ProviderproviderenumNoopenaiopenai, anthropicLLM provider
ModelmodelstringNo""Model name (auto-selected if empty)
Optimization Typeoptimization_typeenumNocomprehensivecomprehensive, performance, cost, scalability, securityType of optimization analysis
FocusfocusstringNo""Specific area to focus on
Include Roadmapinclude_roadmapbooleanNotrueInclude implementation roadmap
Prioritize Quick Winsprioritize_quick_winsbooleanNotruePrioritize low-effort high-impact items

Auth​

SettingNameTypeRequiredDefaultAllowed valuesDescription
API Keyapi_keysecretNoAPI key for the LLM provider
Base URLbase_urlstringNo""Custom API base URL

Advanced​

SettingNameTypeRequiredDefaultAllowed valuesDescription
TemperaturetemperaturenumberNo0.3min 0; max 1Sampling temperature
Max Tokensmax_tokensnumberNo4000min 1Maximum tokens for response

Errors​

CodeDescription
TARGET_REQUIREDNo target input was provided (code, config, or system to optimize)
LLM_REQUEST_FAILEDLLM API call failed during optimization analysis
UNSUPPORTED_PROVIDERProvider must be openai or anthropic

Runtime notes​

  • Each optimization in the array has title, category, priority, effort, risk, and steps fields
  • If the LLM response is not valid JSON, a fallback single-optimization result is returned with the raw text
  • overall_score in metadata is 0-100; it comes from the LLM analysis, not a computed metric
  • The target input can be a string or JSON object; objects are marshaled to JSON before prompting

See Build any workflow for how to put nodes together and Ports, types and data flow for how port types connect.

Quelque chose n'est pas clair ? Envoyez vos commentaires sur la documentation par courriel.