{
  "schemaVersion": 1,
  "updated": "2026-07-12",
  "states": {
    "shipped": {
      "label": "Shipped",
      "description": "Used in a public product or artifact."
    },
    "active": {
      "label": "Active system",
      "description": "Supports a live product or operating workflow."
    },
    "prototype": {
      "label": "Prototype",
      "description": "Implemented in a bounded experiment or proof of concept."
    },
    "historical": {
      "label": "Historical",
      "description": "Previously operated, evaluated, or retained for reference."
    }
  },
  "projects": [
    {
      "id": "trackly",
      "label": "Trackly",
      "color": "#086d55"
    },
    {
      "id": "close-ai",
      "label": "Close AI",
      "color": "#d04f2f"
    },
    {
      "id": "portfolio-labs",
      "label": "Portfolio labs",
      "color": "#3159c7"
    },
    {
      "id": "paypal",
      "label": "PayPal prototype",
      "color": "#7349a8"
    }
  ],
  "categories": [
    {
      "id": "models-agents",
      "label": "Models and agent systems",
      "eyebrow": "Reasoning layer",
      "description": "Models, protocols, and agent runtimes selected for cost, capability, and control."
    },
    {
      "id": "retrieval-data",
      "label": "Retrieval and data",
      "eyebrow": "Context layer",
      "description": "Search, structured data, document parsing, and grounding for product decisions."
    },
    {
      "id": "browser-automation",
      "label": "Browser and automation",
      "eyebrow": "Action layer",
      "description": "Tools and custom policies for workflows that end in real interfaces instead of clean APIs."
    },
    {
      "id": "infrastructure",
      "label": "Infrastructure and delivery",
      "eyebrow": "Runtime layer",
      "description": "Cloud, infrastructure as code, deployment, networking, and production delivery."
    },
    {
      "id": "observability",
      "label": "Observability and analytics",
      "eyebrow": "Learning layer",
      "description": "Tracing, product analytics, model monitoring, alerting, and operational feedback."
    },
    {
      "id": "voice-comms",
      "label": "Voice and communication",
      "eyebrow": "Interaction layer",
      "description": "Real-time voice, transcription, messaging, and conversation-to-action systems."
    },
    {
      "id": "product-quality",
      "label": "Product engineering and quality",
      "eyebrow": "Quality layer",
      "description": "Native and web delivery, automated testing, review, and release evidence."
    },
    {
      "id": "gtm-ops",
      "label": "GTM and operating tools",
      "eyebrow": "Distribution layer",
      "description": "The systems used to reach users, manage relationships, and publish content."
    }
  ],
  "systems": [
    {
      "id": "job-signal",
      "label": "Trackly discovery system",
      "outcome": "Direct career-page signal becomes searchable, explainable job context.",
      "steps": [
        "Puppeteer",
        "PostgreSQL",
        "Azure AI Search",
        "Model routing",
        "Web and native apps"
      ]
    },
    {
      "id": "browser-agent",
      "label": "Human-reviewed browser agent",
      "outcome": "Selected work becomes a prepared form without surrendering submission control.",
      "steps": [
        "Trackly MCP",
        "Custom harness",
        "Codex computer use",
        "Form verification",
        "Human approval"
      ]
    },
    {
      "id": "voice-agent",
      "label": "Real-time voice onboarding",
      "outcome": "A conversation becomes structured profile context inside a native product.",
      "steps": [
        "SwiftUI",
        "Azure Realtime",
        "WebRTC",
        "Tool calls",
        "PostHog and Langfuse"
      ]
    },
    {
      "id": "follow-up",
      "label": "Conversation-to-follow-up",
      "outcome": "Meeting context becomes a reviewed message and a next action.",
      "steps": [
        "Fireflies",
        "Hume AI",
        "Claude",
        "Human review",
        "Gmail and Calendar"
      ]
    }
  ],
  "tools": [
    {
      "id": "anthropic",
      "name": "Anthropic Claude",
      "category": "models-agents",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "icon": "siAnthropic",
      "color": "#b55b38",
      "summary": "Reasoning, generation, and agent workflows across product and operating systems.",
      "proof": "Claude SDK and Agent SDK integrations with routed fallbacks, evaluation, and spend controls.",
      "url": "https://www.anthropic.com/"
    },
    {
      "id": "openai",
      "name": "OpenAI and Codex",
      "category": "models-agents",
      "state": "active",
      "projects": [
        "trackly",
        "portfolio-labs"
      ],
      "color": "#111111",
      "summary": "Model inference, coding agents, and computer-use workflows.",
      "proof": "Used for model calls, Codex-assisted product building, and the human-reviewed browser-agent harness.",
      "url": "https://openai.com/"
    },
    {
      "id": "bedrock",
      "name": "Amazon Bedrock",
      "category": "models-agents",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "color": "#d45b14",
      "summary": "Managed model access inside a resilient inference chain.",
      "proof": "Production fallback tier with routing, cost, and failure behavior documented in the backend.",
      "url": "https://aws.amazon.com/bedrock/"
    },
    {
      "id": "azure-openai",
      "name": "Azure OpenAI",
      "category": "models-agents",
      "state": "active",
      "projects": [
        "trackly"
      ],
      "color": "#0878d1",
      "summary": "Managed inference and a real-time voice agent on Azure.",
      "proof": "Trackly uses short-lived WebRTC sessions for its Azure OpenAI Realtime onboarding agent.",
      "url": "https://azure.microsoft.com/products/ai-services/openai-service"
    },
    {
      "id": "vertex",
      "name": "Google Vertex AI",
      "category": "models-agents",
      "state": "historical",
      "projects": [
        "trackly"
      ],
      "icon": "siGooglecloud",
      "color": "#4285f4",
      "summary": "Evaluated as an additional model tier before routing was consolidated.",
      "proof": "Provider integration remains documented as historical and disabled, not presented as current production.",
      "url": "https://cloud.google.com/vertex-ai"
    },
    {
      "id": "perplexity",
      "name": "Perplexity",
      "category": "models-agents",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "icon": "siPerplexity",
      "color": "#168d8d",
      "summary": "Research and cited web-answer workflows.",
      "proof": "Integrated for research tasks where current web evidence and source attribution matter.",
      "url": "https://www.perplexity.ai/"
    },
    {
      "id": "mcp",
      "name": "Model Context Protocol",
      "category": "models-agents",
      "state": "shipped",
      "projects": [
        "trackly",
        "portfolio-labs",
        "paypal"
      ],
      "icon": "siModelcontextprotocol",
      "color": "#6d49c5",
      "summary": "A public interface that lets AI clients reason over Trackly context.",
      "proof": "Trackly ships both a public CLI MCP server and a hosted OAuth MCP endpoint.",
      "url": "https://github.com/trackly-app/trackly-cli"
    },
    {
      "id": "agent-sdk",
      "name": "Claude Agent SDK",
      "category": "models-agents",
      "state": "prototype",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "icon": "siAnthropic",
      "color": "#b55b38",
      "summary": "Bounded agent execution for multi-step product workflows.",
      "proof": "Used with explicit tool contracts, limits, and fallback behavior rather than an unbounded chat loop.",
      "url": "https://docs.anthropic.com/"
    },
    {
      "id": "postgres",
      "name": "PostgreSQL",
      "category": "retrieval-data",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "icon": "siPostgresql",
      "color": "#32648d",
      "summary": "Primary relational system of record for jobs, users, applications, and operating data.",
      "proof": "The current production database runs on Azure Database for PostgreSQL behind private networking.",
      "url": "https://www.postgresql.org/"
    },
    {
      "id": "azure-search",
      "name": "Azure AI Search",
      "category": "retrieval-data",
      "state": "active",
      "projects": [
        "trackly"
      ],
      "color": "#0878d1",
      "summary": "Hybrid keyword and vector retrieval for job discovery.",
      "proof": "Production search combines semantic retrieval with SQL fallbacks and query telemetry.",
      "url": "https://azure.microsoft.com/products/ai-services/ai-search"
    },
    {
      "id": "llamaindex",
      "name": "LlamaIndex",
      "category": "retrieval-data",
      "state": "prototype",
      "projects": [
        "trackly"
      ],
      "color": "#7b4ff2",
      "summary": "Document parsing and extraction experiments for resume context.",
      "proof": "Evaluated for turning uploaded PDF resumes into structured product context.",
      "url": "https://www.llamaindex.ai/"
    },
    {
      "id": "exa",
      "name": "Exa",
      "category": "retrieval-data",
      "state": "prototype",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "color": "#f05f43",
      "summary": "AI-native web discovery for research-heavy workflows.",
      "proof": "Evaluated in hiring-manager and company research flows where semantic web discovery is useful.",
      "url": "https://exa.ai/"
    },
    {
      "id": "tavily",
      "name": "Tavily",
      "category": "retrieval-data",
      "state": "prototype",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "color": "#24957a",
      "summary": "Search API for grounded agent research.",
      "proof": "Configured as a research provider for public hiring signals and source-backed enrichment.",
      "url": "https://tavily.com/"
    },
    {
      "id": "parallel",
      "name": "Parallel",
      "category": "retrieval-data",
      "state": "prototype",
      "projects": [
        "trackly"
      ],
      "color": "#111111",
      "summary": "Evaluated for deep web research and extraction workloads.",
      "proof": "Used in research experiments; not represented as a production dependency.",
      "url": "https://parallel.ai/"
    },
    {
      "id": "computer-use",
      "name": "Codex computer use",
      "category": "browser-automation",
      "state": "prototype",
      "projects": [
        "trackly"
      ],
      "color": "#111111",
      "summary": "Chrome form preparation grounded in Trackly job and profile context.",
      "proof": "The agent fills and verifies fields, then pauses before submission for fresh human approval.",
      "url": "https://portfolio.kevinastuhuaman.com/projects/trackly/"
    },
    {
      "id": "agent-harness",
      "name": "Custom agent harness",
      "category": "browser-automation",
      "state": "prototype",
      "projects": [
        "trackly"
      ],
      "color": "#086d55",
      "summary": "Domain rules around a general computer-use model.",
      "proof": "Encodes ATS mechanics, committed form state, recovery, approval invalidation, and confirmation handling.",
      "url": "https://portfolio.kevinastuhuaman.com/projects/trackly/"
    },
    {
      "id": "puppeteer",
      "name": "Puppeteer",
      "category": "browser-automation",
      "state": "active",
      "projects": [
        "trackly"
      ],
      "icon": "siPuppeteer",
      "color": "#159b77",
      "summary": "Browser automation for source collection and structured web workflows.",
      "proof": "Used in the backend for browser-based ATS coverage and controlled automation.",
      "url": "https://pptr.dev/"
    },
    {
      "id": "playwright",
      "name": "Playwright",
      "category": "browser-automation",
      "state": "active",
      "projects": [
        "trackly",
        "portfolio-labs"
      ],
      "color": "#397d45",
      "summary": "Browser-level product verification across responsive states.",
      "proof": "Runs end-to-end tests and visual checks for Trackly Web and public product studies.",
      "url": "https://playwright.dev/"
    },
    {
      "id": "cdp",
      "name": "Chrome DevTools Protocol",
      "category": "browser-automation",
      "state": "shipped",
      "projects": [
        "close-ai",
        "portfolio-labs"
      ],
      "icon": "siGooglechrome",
      "color": "#4285f4",
      "summary": "Direct browser control for publishing and repeatable operations.",
      "proof": "Used in a public X article publisher and browser-control workflows.",
      "url": "https://chromedevtools.github.io/devtools-protocol/"
    },
    {
      "id": "human-approval",
      "name": "Human approval gates",
      "category": "browser-automation",
      "state": "shipped",
      "projects": [
        "trackly",
        "close-ai",
        "portfolio-labs"
      ],
      "color": "#d04f2f",
      "summary": "Product policy for irreversible or subjective actions.",
      "proof": "Applied to browser submission, generated follow-ups, and evidence-aware investigation patterns.",
      "url": "https://kevinastuhuaman.github.io/ai-investigation-workbench/"
    },
    {
      "id": "azure",
      "name": "Microsoft Azure",
      "category": "infrastructure",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "color": "#0878d1",
      "summary": "Current application, worker, search, networking, and database platform.",
      "proof": "The live Trackly backend and data plane run on Azure after a documented cloud cutover.",
      "url": "https://azure.microsoft.com/"
    },
    {
      "id": "azure-app-service",
      "name": "Azure App Service",
      "category": "infrastructure",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "color": "#0878d1",
      "summary": "Managed runtime for public web, API, worker, and monitoring surfaces.",
      "proof": "The current Azure deployment separates user-facing, API, and background workloads with guarded CI paths.",
      "url": "https://azure.microsoft.com/products/app-service"
    },
    {
      "id": "azure-key-vault",
      "name": "Azure Key Vault",
      "category": "infrastructure",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "color": "#0878d1",
      "summary": "Managed secret and key boundary for the Azure operating environment.",
      "proof": "Included in the reviewed production infrastructure inventory without publishing values or access topology.",
      "url": "https://azure.microsoft.com/products/key-vault"
    },
    {
      "id": "terraform",
      "name": "Terraform",
      "category": "infrastructure",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "icon": "siTerraform",
      "color": "#7046b2",
      "summary": "Audit-first infrastructure documentation and drift detection.",
      "proof": "Azure production and AWS rollback resources have reviewed, non-mutating Terraform plans.",
      "url": "https://www.terraform.io/"
    },
    {
      "id": "aws-lightsail",
      "name": "AWS Lightsail",
      "category": "infrastructure",
      "state": "historical",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "color": "#d45b14",
      "summary": "Previous application and worker hosting retained as rollback reference.",
      "proof": "Modeled in the cloud migration inventory as historical infrastructure, not current production.",
      "url": "https://aws.amazon.com/lightsail/"
    },
    {
      "id": "aws-rds",
      "name": "Amazon RDS",
      "category": "infrastructure",
      "state": "historical",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "color": "#d45b14",
      "summary": "Previous managed PostgreSQL platform retained for rollback reference.",
      "proof": "Documented in the audited AWS-to-Azure migration rather than presented as live production.",
      "url": "https://aws.amazon.com/rds/"
    },
    {
      "id": "aws-ses",
      "name": "Amazon SES",
      "category": "infrastructure",
      "state": "active",
      "projects": [
        "close-ai"
      ],
      "color": "#d45b14",
      "summary": "Transactional email delivery for communication workflows.",
      "proof": "Used behind email generation and follow-up operations with deliverability controls.",
      "url": "https://aws.amazon.com/ses/"
    },
    {
      "id": "cloudflare",
      "name": "Cloudflare",
      "category": "infrastructure",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai",
        "portfolio-labs"
      ],
      "icon": "siCloudflare",
      "color": "#f48120",
      "summary": "DNS, edge routing, TLS, and public-surface controls.",
      "proof": "Supports Trackly domains and the recruiter portfolio delivery path.",
      "url": "https://www.cloudflare.com/"
    },
    {
      "id": "docker",
      "name": "Docker",
      "category": "infrastructure",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai",
        "portfolio-labs"
      ],
      "icon": "siDocker",
      "color": "#2496ed",
      "summary": "Repeatable service packaging and local system simulation.",
      "proof": "Used for backend delivery and multi-service public architecture demonstrations.",
      "url": "https://www.docker.com/"
    },
    {
      "id": "github-actions",
      "name": "GitHub Actions",
      "category": "infrastructure",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai",
        "portfolio-labs"
      ],
      "icon": "siGithubactions",
      "color": "#2088ff",
      "summary": "Build, review, security, deployment, and verification automation.",
      "proof": "Runs multi-repo CI and the guarded deployment paths for web, backend, native, and public labs.",
      "url": "https://github.com/features/actions"
    },
    {
      "id": "vercel",
      "name": "Vercel",
      "category": "infrastructure",
      "state": "historical",
      "projects": [
        "trackly",
        "portfolio-labs"
      ],
      "icon": "siVercel",
      "color": "#111111",
      "summary": "Previous web deployment and fast prototype hosting.",
      "proof": "Used for earlier product surfaces before the current Azure and GitHub Pages paths.",
      "url": "https://vercel.com/"
    },
    {
      "id": "railway",
      "name": "Railway",
      "category": "infrastructure",
      "state": "historical",
      "projects": [
        "close-ai"
      ],
      "icon": "siRailway",
      "color": "#6f3ed8",
      "summary": "Early managed deployment platform for rapid product iteration.",
      "proof": "Retained as historical builder experience, not current infrastructure.",
      "url": "https://railway.com/"
    },
    {
      "id": "langfuse",
      "name": "Langfuse",
      "category": "observability",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "color": "#111111",
      "summary": "Self-hosted LLM tracing, usage, and evaluation telemetry.",
      "proof": "Instrumented across model routes with graceful failure behavior and public architecture notes.",
      "url": "https://langfuse.com/"
    },
    {
      "id": "posthog",
      "name": "PostHog",
      "category": "observability",
      "state": "active",
      "projects": [
        "trackly"
      ],
      "icon": "siPosthog",
      "color": "#e8590c",
      "summary": "Product analytics for web and native behavior.",
      "proof": "Tracks funnels and product decisions across Trackly surfaces with privacy-aware event design.",
      "url": "https://posthog.com/"
    },
    {
      "id": "umami",
      "name": "Umami",
      "category": "observability",
      "state": "shipped",
      "projects": [
        "portfolio-labs",
        "close-ai"
      ],
      "icon": "siUmami",
      "color": "#111111",
      "summary": "Privacy-conscious analytics exposed to AI through MCP.",
      "proof": "A public Go MCP server turns Umami analytics into structured assistant tools.",
      "url": "https://github.com/kevinastuhuaman/umami-mcp-server"
    },
    {
      "id": "gatus",
      "name": "Gatus",
      "category": "observability",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "color": "#3159c7",
      "summary": "Public endpoint health checks and availability monitoring.",
      "proof": "Deployed as part of the current public-surface monitoring path.",
      "url": "https://github.com/TwiN/gatus"
    },
    {
      "id": "healthchecks",
      "name": "Healthchecks",
      "category": "observability",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "color": "#3a9b55",
      "summary": "Heartbeat monitoring for scheduled and background work.",
      "proof": "Used to detect silent failures in recurring operational workflows.",
      "url": "https://healthchecks.io/"
    },
    {
      "id": "firebase",
      "name": "Firebase",
      "category": "observability",
      "state": "active",
      "projects": [
        "trackly"
      ],
      "icon": "siFirebase",
      "color": "#ff9e0b",
      "summary": "Native crash reporting, remote configuration, and app telemetry.",
      "proof": "Integrated into Trackly iOS and macOS release builds.",
      "url": "https://firebase.google.com/"
    },
    {
      "id": "datadog",
      "name": "Datadog",
      "category": "observability",
      "state": "prototype",
      "projects": [
        "paypal"
      ],
      "icon": "siDatadog",
      "color": "#6c3aa8",
      "summary": "Diagnostics and telemetry inside an enterprise observability prototype.",
      "proof": "Used during the PayPal Checkout internship prototype; no employer data or architecture is published.",
      "url": "https://www.datadoghq.com/"
    },
    {
      "id": "cloudwatch",
      "name": "Amazon CloudWatch",
      "category": "observability",
      "state": "historical",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "icon": "siAmazoncloudwatch",
      "color": "#d45b14",
      "summary": "Infrastructure logging and operational visibility on the prior AWS stack.",
      "proof": "Documented as experience from the AWS operating period, now secondary to Azure monitoring.",
      "url": "https://aws.amazon.com/cloudwatch/"
    },
    {
      "id": "application-insights",
      "name": "Application Insights",
      "category": "observability",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "color": "#0878d1",
      "summary": "Application performance and failure telemetry on the current Azure runtime.",
      "proof": "Monitored as part of the production public-surface and infrastructure drift checks.",
      "url": "https://learn.microsoft.com/azure/azure-monitor/app/app-insights-overview"
    },
    {
      "id": "log-analytics",
      "name": "Azure Log Analytics",
      "category": "observability",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "color": "#0878d1",
      "summary": "Centralized operational logs for Azure-hosted services.",
      "proof": "Included in the current observability inventory while keys and private workspace details remain excluded.",
      "url": "https://learn.microsoft.com/azure/azure-monitor/logs/log-analytics-overview"
    },
    {
      "id": "azure-realtime",
      "name": "Azure OpenAI Realtime",
      "category": "voice-comms",
      "state": "shipped",
      "projects": [
        "trackly"
      ],
      "color": "#0878d1",
      "summary": "A low-latency voice onboarding agent over WebRTC.",
      "proof": "The native apps use ephemeral tokens, direct audio transport, tool calls, and session observability.",
      "url": "https://learn.microsoft.com/azure/ai-foundry/openai/how-to/realtime-audio-webrtc"
    },
    {
      "id": "webrtc",
      "name": "WebRTC",
      "category": "voice-comms",
      "state": "shipped",
      "projects": [
        "trackly"
      ],
      "icon": "siWebrtc",
      "color": "#3159c7",
      "summary": "Real-time audio transport for Trackly voice interactions.",
      "proof": "Implemented in Swift with reconnect handling, audio routing, and sanitized transport metrics.",
      "url": "https://webrtc.org/"
    },
    {
      "id": "whisper",
      "name": "Whisper",
      "category": "voice-comms",
      "state": "shipped",
      "projects": [
        "trackly"
      ],
      "color": "#111111",
      "summary": "On-device speech transcription for native voice input.",
      "proof": "Trackly includes a Whisper-based transcription path in its iOS product.",
      "url": "https://github.com/openai/whisper"
    },
    {
      "id": "fireflies",
      "name": "Fireflies",
      "category": "voice-comms",
      "state": "active",
      "projects": [
        "close-ai"
      ],
      "color": "#7046b2",
      "summary": "Conversation capture that triggers follow-up workflows.",
      "proof": "Webhook integrations turn completed conversations into structured drafting tasks.",
      "url": "https://fireflies.ai/"
    },
    {
      "id": "hume",
      "name": "Hume AI",
      "category": "voice-comms",
      "state": "prototype",
      "projects": [
        "close-ai"
      ],
      "color": "#d04f2f",
      "summary": "Voice emotion analysis for conversation context.",
      "proof": "Used to test whether emotional signals can improve follow-up quality and recall.",
      "url": "https://www.hume.ai/"
    },
    {
      "id": "twilio",
      "name": "Twilio",
      "category": "voice-comms",
      "state": "active",
      "projects": [
        "close-ai"
      ],
      "color": "#f22f46",
      "summary": "SMS notifications for approval and operating workflows.",
      "proof": "Connects background processing to time-sensitive human review.",
      "url": "https://www.twilio.com/"
    },
    {
      "id": "vapi",
      "name": "Vapi",
      "category": "voice-comms",
      "state": "historical",
      "projects": [
        "trackly",
        "close-ai"
      ],
      "color": "#111111",
      "summary": "Earlier voice-agent prototyping before the direct Azure Realtime path.",
      "proof": "Kept as historical product exploration rather than presented as the current implementation.",
      "url": "https://vapi.ai/"
    },
    {
      "id": "nextjs",
      "name": "Next.js and React",
      "category": "product-quality",
      "state": "shipped",
      "projects": [
        "trackly"
      ],
      "icon": "siNextdotjs",
      "color": "#111111",
      "summary": "The production web product and recruiter-facing job workflows.",
      "proof": "Trackly Web ships responsive product surfaces, agent chat, onboarding, and job management.",
      "url": "https://nextjs.org/"
    },
    {
      "id": "swiftui",
      "name": "SwiftUI",
      "category": "product-quality",
      "state": "shipped",
      "projects": [
        "trackly"
      ],
      "icon": "siSwift",
      "color": "#f05138",
      "summary": "Native iOS and macOS product delivery.",
      "proof": "Trackly ships separate native applications with shared product contracts and platform-specific UX.",
      "url": "https://developer.apple.com/xcode/swiftui/"
    },
    {
      "id": "astro",
      "name": "Astro",
      "category": "product-quality",
      "state": "shipped",
      "projects": [
        "portfolio-labs"
      ],
      "icon": "siAstro",
      "color": "#ff5d01",
      "summary": "Fast static delivery for interactive public product studies.",
      "proof": "Powers the investigation workbench, recruiter portfolio, and this evidence map.",
      "url": "https://astro.build/"
    },
    {
      "id": "typescript",
      "name": "TypeScript",
      "category": "product-quality",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai",
        "portfolio-labs"
      ],
      "icon": "siTypescript",
      "color": "#3178c6",
      "summary": "Shared implementation language for web, backend, tests, and public demos.",
      "proof": "Used across production services and public-safe prototypes with automated type checks.",
      "url": "https://www.typescriptlang.org/"
    },
    {
      "id": "go",
      "name": "Go",
      "category": "product-quality",
      "state": "shipped",
      "projects": [
        "portfolio-labs"
      ],
      "icon": "siGo",
      "color": "#00add8",
      "summary": "Compact systems implementation for the Umami MCP server.",
      "proof": "The public server includes tests, schemas, container packaging, and release automation.",
      "url": "https://go.dev/"
    },
    {
      "id": "coderabbit",
      "name": "CodeRabbit",
      "category": "product-quality",
      "state": "active",
      "projects": [
        "trackly",
        "close-ai",
        "portfolio-labs"
      ],
      "icon": "siCoderabbit",
      "color": "#e8590c",
      "summary": "Automated review inside a multi-reviewer shipping gate.",
      "proof": "Used with CI, Codex, security review, and explicit resolution rather than treated as a sole approval.",
      "url": "https://coderabbit.ai/"
    },
    {
      "id": "conductor",
      "name": "Conductor",
      "category": "product-quality",
      "state": "active",
      "projects": [
        "trackly",
        "portfolio-labs"
      ],
      "color": "#3159c7",
      "summary": "Parallel coding workspaces for isolated product delivery.",
      "proof": "Used to coordinate multiple bounded branches while preserving review and merge gates.",
      "url": "https://www.conductor.build/"
    },
    {
      "id": "figma",
      "name": "Figma",
      "category": "product-quality",
      "state": "prototype",
      "projects": [
        "portfolio-labs"
      ],
      "icon": "siFigma",
      "color": "#a43d22",
      "summary": "High-fidelity product and interaction prototyping.",
      "proof": "Used for Berkeley and MoBagel AI product concepts, UX exploration, and roadmap communication.",
      "url": "https://www.figma.com/"
    },
    {
      "id": "remotion",
      "name": "Remotion",
      "category": "product-quality",
      "state": "shipped",
      "projects": [
        "trackly",
        "portfolio-labs"
      ],
      "icon": "siRemotion",
      "color": "#3159c7",
      "summary": "Programmatic motion graphics for product storytelling.",
      "proof": "Used to ship a public Trackly CLI product video with repeatable, code-reviewed motion.",
      "url": "https://www.remotion.dev/"
    },
    {
      "id": "xcode",
      "name": "Xcode and TestFlight",
      "category": "product-quality",
      "state": "shipped",
      "projects": [
        "trackly"
      ],
      "icon": "siXcode",
      "color": "#0878d1",
      "summary": "Native build, signing, beta distribution, and release verification.",
      "proof": "Used for Trackly iOS and macOS delivery, including guarded builds and TestFlight distribution.",
      "url": "https://developer.apple.com/testflight/"
    },
    {
      "id": "replit",
      "name": "Replit",
      "category": "product-quality",
      "state": "historical",
      "projects": [
        "portfolio-labs",
        "close-ai"
      ],
      "icon": "siReplit",
      "color": "#a43d22",
      "summary": "Fast collaborative prototyping for earlier product experiments.",
      "proof": "Included as historical builder experience rather than a current delivery dependency.",
      "url": "https://replit.com/"
    },
    {
      "id": "clay",
      "name": "Clay",
      "category": "gtm-ops",
      "state": "active",
      "projects": [
        "close-ai"
      ],
      "color": "#e36b43",
      "summary": "Contact enrichment and structured GTM research.",
      "proof": "Used in relationship and outreach workflows with explicit enrichment quality checks.",
      "url": "https://www.clay.com/"
    },
    {
      "id": "hubspot",
      "name": "HubSpot",
      "category": "gtm-ops",
      "state": "active",
      "projects": [
        "close-ai"
      ],
      "icon": "siHubspot",
      "color": "#ff5c35",
      "summary": "CRM integration for contact and relationship operations.",
      "proof": "Used for contact management and source attribution workflows.",
      "url": "https://www.hubspot.com/"
    },
    {
      "id": "ghost",
      "name": "Ghost",
      "category": "gtm-ops",
      "state": "shipped",
      "projects": [
        "close-ai",
        "portfolio-labs"
      ],
      "icon": "siGhost",
      "color": "#111111",
      "summary": "Newsletter publishing and owned-audience distribution.",
      "proof": "Used in a voice-to-newsletter workflow and Kevin's publishing system.",
      "url": "https://ghost.org/"
    },
    {
      "id": "gmail",
      "name": "Gmail and Calendar",
      "category": "gtm-ops",
      "state": "active",
      "projects": [
        "close-ai"
      ],
      "icon": "siGmail",
      "color": "#4285f4",
      "summary": "Draft, send, schedule, and follow-up actions tied to conversation context.",
      "proof": "Integrated through OAuth with approval-oriented relationship workflows.",
      "url": "https://workspace.google.com/"
    },
    {
      "id": "notion",
      "name": "Notion",
      "category": "gtm-ops",
      "state": "active",
      "projects": [
        "close-ai"
      ],
      "icon": "siNotion",
      "color": "#111111",
      "summary": "Structured operating context and relationship data.",
      "proof": "Used in integration and workflow experiments where editable human context matters.",
      "url": "https://www.notion.so/"
    }
  ]
}