// Skills

The toolkit I use to
ship real systems

Battle-tested in production — not just on paper.

AI & LLM

OpenAI APIAnthropic ClaudeGoogle GeminiClaude CodeAI AgentsChatbotsVoice & Call AgentsFunction / Tool CallingAgentic WorkflowsPrompt EngineeringConversation MemoryHuman-in-the-Loop

RAG & Vector Search

RAG ArchitecturesEmbeddingsPineconepgvectorChromaLangChainLangGraphLlamaIndexChunking StrategiesSemantic SearchRerankingCustom MCP Servers

Automation & Integrations

n8nMake.comZapierPower AutomateGoHighLevelWebhooksREST API IntegrationOAuth 2.0Scheduled JobsWhatsApp / Meta Cloud APITwilioPlaywright Scraping

Backend & APIs

PythonFastAPINode.jsNestJSExpress.jsTypeScriptREST APIsGraphQLWebSocketsMicroservicesJWT AuthRBAC

Frontend

ReactNext.jsTypeScriptJavaScriptTailwind CSSResponsive UIREST / HTTPFramer Motion

Databases

PostgreSQLMongoDBRedisPrisma ORMData ModelingQuery OptimizationCachingETL

DevOps & Cloud

DockerKubernetesHelmCI/CDGitHub ActionsNginxDigitalOceanAWSLinuxBash

Monitoring & QA

PrometheusGrafanaLog ManagementAlerts & DashboardsHealth ChecksJestUnit TestingCode Review

Languages & CP

PythonTypeScriptJavaScriptC++SQLData StructuresAlgorithmsDynamic Programming
// AI & Automation

AI agents
wired into real
backend systems.

I build the AI layer that sits on top of solid backends — agents, LLM pipelines, custom MCP servers, and workflows that call real tools, take actions, and run reliably in production.

AI Agents

Chat, Voice & Call Agents

I build AI agents that plan steps, use tools, remember context, and fix errors on their own — chat, voice, and full call agents that talk to customers in real time.

  • Lead bots that talk to customers 24/7
  • Voice and call agents with OpenAI Realtime
  • Tool-calling agents that take real actions
  • Agents that read Slack and write reports
LLM Integration

OpenAI & Claude in Production

I integrate OpenAI and Claude into real backends — RAG over your data, function calling, and reliable prompt pipelines that run at scale, not just demos.

  • RAG over private data and docs
  • Function calling and structured outputs
  • Streaming responses and token control
  • Guardrails and fallback handling
MCP & Claude Code

Custom MCP Servers & Agentic Ops

I build agentic systems with Claude Code — custom MCP servers, Slack and GitHub integrations, and AI workflows that ship real code and run real operations.

  • Custom MCP servers for client tools
  • Claude Code agents for code review and ops
  • AI-driven CI/CD and deploy helpers
  • Internal tools that read context and act
Automation

n8n & Make Workflows

I build custom n8n and Make.com flows that connect AI agents to your business apps — easy for non-tech people to use and edit later.

  • Custom n8n nodes for any API
  • AI nodes for OpenAI and Claude
  • Self-hosted n8n on Docker
  • Make.com scenarios with branching logic

Stack & Tooling

The AI, automation, and engineering tools I run in production

16 active
Node.jsCore
NestJSAPIs
Express.jsAPIs
PostgreSQLLive
MongoDBLive
RedisCache
DockerContainers
KubernetesOrchestration
GitHub ActionsCI/CD
PrometheusMetrics
GrafanaDashboards
DigitalOceanCloud
OpenAI GPT-4Production
Anthropic ClaudeProduction
Claude Code MCPCustom
n8n Self-HostedDocker
Build an AI System With Me