-
Notifications
You must be signed in to change notification settings - Fork 0
Expand file tree
/
Copy pathAIML-Stacks-Visual-Guide.html
More file actions
648 lines (599 loc) · 50.4 KB
/
Copy pathAIML-Stacks-Visual-Guide.html
File metadata and controls
648 lines (599 loc) · 50.4 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
<!doctype html>
<html lang="en">
<head>
<meta charset="utf-8" />
<meta name="viewport" content="width=device-width,initial-scale=1" />
<title>AI/ML Stacks · Interactive Visual Guide 2026</title>
<meta name="description" content="Interactive visual guide to AI/ML tech stacks in 2026. Compare LLM providers, frameworks, vector databases, RAG, agents, fine-tuning, eval, and more." />
<meta name="keywords" content="AI, ML, tech stack, 2026, LLM, RAG, vector database, agents, fine-tuning, OpenAI, Claude, LangChain" />
<meta property="og:title" content="AI/ML Tech Stacks · Visual Guide 2026" />
<meta property="og:description" content="Mind-map view of the 2026 AI/ML universe — LLM providers, frameworks, vector DBs, RAG, agents, fine-tuning, eval, inference, and data." />
<meta property="og:type" content="website" />
<meta property="og:url" content="https://opentechstack.stitchwebsite.com/AIML-Stacks-Visual-Guide.html" />
<meta name="twitter:card" content="summary_large_image" />
<meta name="twitter:title" content="AI/ML Tech Stacks · Visual Guide 2026" />
<meta name="twitter:description" content="Interactive guide to the 2026 AI/ML stack landscape. Compare 40+ technologies across 10 categories." />
<meta name="robots" content="index, follow, max-snippet:-1, max-image-preview:large" />
<link rel="canonical" href="https://opentechstack.stitchwebsite.com/AIML-Stacks-Visual-Guide.html" />
<script type="application/ld+json">
{"@context":"https://schema.org","@type":"Article","name":"AI/ML Stacks Visual Guide 2026","headline":"AI/ML Tech Stacks — Interactive Visual Guide 2026","description":"Interactive guide to 45+ AI/ML technologies: OpenAI, Claude, Gemini, LangChain, LlamaIndex, Pinecone, CrewAI, vLLM, and more.","url":"https://opentechstack.stitchwebsite.com/AIML-Stacks-Visual-Guide.html","publisher":{"@type":"Organization","name":"Open TechStack","url":"https://opentechstack.stitchwebsite.com"},"datePublished":"2026-04-06","dateModified":"2026-04-06","keywords":"AI, ML, LLM, RAG, vector database, agents, OpenAI, Claude, LangChain, tech stack, 2026","isPartOf":{"@type":"WebSite","name":"Open TechStack","url":"https://opentechstack.stitchwebsite.com"}}
</script>
<link rel="icon" href="data:image/svg+xml,<svg xmlns='http://www.w3.org/2000/svg' viewBox='0 0 100 100'><text y='.9em' font-size='90'>🧠</text></svg>" />
<script src="https://cdn.tailwindcss.com"></script>
<script>tailwind.config={darkMode:"class"}</script>
<link rel="stylesheet" href="dark-mode.css">
<script src="dark-mode.js"></script>
<style>
html { scroll-behavior: smooth; }
body { font-family: ui-sans-serif, system-ui, -apple-system, "Segoe UI", Roboto, sans-serif; }
.cat-llm { --c:#6366f1; --cs:#eef2ff; } /* LLM Providers */
.cat-fw { --c:#06b6d4; --cs:#ecfeff; } /* Frameworks & Orchestration */
.cat-vec { --c:#10b981; --cs:#ecfdf5; } /* Vector Databases */
.cat-embed { --c:#8b5cf6; --cs:#f5f3ff; } /* Embeddings & Models */
.cat-rag { --c:#f59e0b; --cs:#fffbeb; } /* RAG & Retrieval */
.cat-agent { --c:#e11d48; --cs:#fff1f2; } /* Agents & Tools */
.cat-fine { --c:#ec4899; --cs:#fdf2f8; } /* Fine-tuning & Training */
.cat-eval { --c:#0ea5e9; --cs:#f0f9ff; } /* Eval & Monitoring */
.cat-infra { --c:#14b8a6; --cs:#f0fdfa; } /* Inference & Infra */
.cat-data { --c:#f97316; --cs:#fff7ed; } /* Data & Labeling */
.chip { background: var(--cs); color: var(--c); border: 1px solid color-mix(in srgb, var(--c) 25%, transparent); }
.node { cursor: pointer; transition: transform .2s ease, opacity .3s ease; transform-origin: center; transform-box: fill-box; }
.node:hover { transform: scale(1.08); }
.node.dim { opacity: .18; }
.link { transition: opacity .3s ease, stroke-width .2s ease; }
.link.dim { opacity: .1; }
.card { transition: opacity .25s ease, transform .25s ease, box-shadow .25s ease; }
.card.hidden-card { opacity: 0; transform: scale(.96); pointer-events: none; position: absolute; }
.card.flash { animation: flash 1.2s ease; }
@keyframes flash {
0%,100% { box-shadow: 0 1px 2px rgba(0,0,0,.05); }
30% { box-shadow: 0 0 0 4px #fde68a, 0 10px 30px rgba(245,158,11,.25); }
}
.stage-btn.active { background: #0f172a; color: white; box-shadow: 0 8px 20px rgba(15,23,42,.25); }
.filter-chip.active { background: #0f172a !important; color: white !important; border-color: #0f172a !important; }
@media print {
.no-print { display: none !important; }
.card.hidden-card { display: none !important; }
}
</style>
</head>
<body class="bg-slate-50 text-slate-800">
<!-- ============ NAV ============ -->
<nav class="bg-slate-900 text-white text-sm sticky top-0 z-50">
<div class="max-w-5xl mx-auto px-6 py-2.5 flex items-center justify-between">
<a href="index.html" class="font-bold tracking-tight hover:text-slate-300 transition">⚡ Open TechStack</a>
<div class="flex items-center gap-4 flex-wrap">
<a href="Frontend-Stacks-Visual-Guide.html" class="hover:text-slate-300 transition">Frontend</a>
<a href="Backend-Stacks-Visual-Guide.html" class="hover:text-slate-300 transition">Backend</a>
<a href="AIML-Stacks-Visual-Guide.html" class="hover:text-slate-300 transition">AI/ML</a>
<a href="AI-Coding-Tools-Visual-Guide.html" class="hover:text-slate-300 transition">AI Coding</a>
<a href="DevOps-Stacks-Visual-Guide.html" class="hover:text-slate-300 transition">DevOps</a>
<a href="Mobile-Stacks-Visual-Guide.html" class="hover:text-slate-300 transition">Mobile</a>
<a href="Stack-Combos.html" class="hover:text-slate-300 transition">Combos</a>
<a href="stack-picker.html" class="hover:text-slate-300 transition">Picker</a>
<a href="stack-detector.html" class="hover:text-slate-300 transition text-emerald-400 font-semibold">Detector</a>
<a href="blog/" class="hover:text-slate-300 transition">Blog</a>
<a href="compare/" class="hover:text-slate-300 transition">Compare</a>
<a href="https://github.com/JagadeepPortfolio/awesome-vibecoding-techstacks" target="_blank" rel="noopener" class="hover:text-slate-300 transition">GitHub</a>
</div>
</div>
</nav>
<header class="relative overflow-hidden border-b border-slate-200 bg-gradient-to-br from-white via-violet-50 to-emerald-50">
<div class="max-w-7xl mx-auto px-6 py-10">
<div class="flex items-start justify-between gap-6 flex-wrap">
<div class="max-w-3xl">
<div class="inline-flex items-center gap-2 text-xs font-medium text-violet-700 bg-violet-100 rounded-full px-3 py-1 mb-3">
<span class="w-2 h-2 rounded-full bg-violet-500"></span>
Interactive guide · 2026 edition
</div>
<h1 class="text-4xl md:text-5xl font-extrabold tracking-tight text-slate-900">AI/ML Tech Stacks, Visualised</h1>
<p class="mt-3 text-slate-600 text-lg">
A mind-map view of the 2026 AI/ML universe — LLM providers, frameworks, vector databases, RAG, agents, fine-tuning, eval, inference, and data.
Pick your stage, filter by what matters, and see the stack that fits.
</p>
</div>
<div class="no-print flex items-center gap-2 text-xs text-slate-500">
<kbd class="px-2 py-1 bg-white border border-slate-300 rounded">/</kbd> search
<kbd class="px-2 py-1 bg-white border border-slate-300 rounded">1-4</kbd> stage
<kbd class="px-2 py-1 bg-white border border-slate-300 rounded">Esc</kbd> reset
</div>
</div>
</div>
</header>
<section class="no-print sticky top-0 z-30 bg-white/90 backdrop-blur border-b border-slate-200 shadow-sm">
<div class="max-w-7xl mx-auto px-6 py-4 flex flex-col gap-3">
<div class="flex items-center gap-3 flex-wrap">
<div class="relative flex-1 min-w-[220px]">
<svg class="w-4 h-4 absolute left-3 top-1/2 -translate-y-1/2 text-slate-400" viewBox="0 0 24 24" fill="none" stroke="currentColor" stroke-width="2"><circle cx="11" cy="11" r="8"/><path d="m21 21-4.35-4.35"/></svg>
<input id="search" type="text" placeholder="Search LLM, framework, vector DB, or tool…"
class="w-full pl-9 pr-3 py-2 rounded-lg border border-slate-300 focus:outline-none focus:ring-2 focus:ring-violet-400 bg-white text-sm" />
</div>
<label class="inline-flex items-center gap-2 text-sm text-slate-600"><input type="checkbox" id="f-free" class="rounded border-slate-300"> Free tier</label>
<label class="inline-flex items-center gap-2 text-sm text-slate-600"><input type="checkbox" id="f-oss" class="rounded border-slate-300"> Open source</label>
<label class="inline-flex items-center gap-2 text-sm text-slate-600"><input type="checkbox" id="f-prod" class="rounded border-slate-300"> Production-proven</label>
<button id="reset" class="ml-auto text-xs text-slate-500 hover:text-slate-900 underline">Reset all</button>
</div>
<div id="catChips" class="flex items-center gap-2 flex-wrap text-xs"></div>
</div>
</section>
<section class="max-w-7xl mx-auto px-6 pt-8">
<div class="flex items-center justify-between mb-3">
<h2 class="text-lg font-semibold text-slate-900">The landscape at a glance</h2>
<p class="text-xs text-slate-500"><span id="countLabel">—</span> · click a node to jump to its card</p>
</div>
<div class="rounded-2xl border border-slate-200 bg-white shadow-sm overflow-hidden">
<svg id="mindmap" viewBox="0 0 1200 760" class="w-full h-[560px]"></svg>
</div>
</section>
<section class="max-w-7xl mx-auto px-6 py-10">
<div class="flex items-center justify-between mb-4 flex-wrap gap-3">
<div>
<h2 class="text-lg font-semibold text-slate-900">Which stage are you at?</h2>
<p class="text-sm text-slate-500">Pick one — the stack below adapts, and cards fade to only what fits.</p>
</div>
<div id="stageTabs" class="flex gap-2 flex-wrap"></div>
</div>
<div id="stagePanel" class="rounded-2xl border border-slate-200 bg-gradient-to-br from-white to-slate-50 shadow-sm p-6"></div>
</section>
<section class="max-w-7xl mx-auto px-6 pb-10">
<div class="flex items-center justify-between mb-4">
<h2 class="text-lg font-semibold text-slate-900">All technologies</h2>
<p class="text-xs text-slate-500" id="gridCount"></p>
</div>
<div id="grid" class="grid grid-cols-1 md:grid-cols-2 lg:grid-cols-3 gap-4"></div>
</section>
<section class="bg-white border-t border-slate-200">
<div class="max-w-7xl mx-auto px-6 py-12">
<h2 class="text-xl font-bold text-slate-900 mb-6">Decision cheat-sheet</h2>
<div class="grid grid-cols-1 md:grid-cols-3 gap-6">
<div class="rounded-xl border border-slate-200 p-5 bg-slate-50">
<h3 class="font-semibold text-slate-900 mb-3">Pick by use case</h3>
<ul class="space-y-2 text-sm text-slate-700">
<li><b>Chatbot:</b> OpenAI / Claude + Vercel AI SDK. Streaming-first, great UX out of the box.</li>
<li><b>RAG app:</b> LlamaIndex + Pinecone + Claude. Best data connectors and retrieval quality.</li>
<li><b>AI agents:</b> LangGraph + Claude MCP. Stateful multi-step workflows with tool access.</li>
<li><b>Code assistant:</b> Claude + fine-tuned model. Best-in-class code generation and analysis.</li>
<li><b>Search:</b> Cohere Embed + Qdrant. Top embeddings with high-performance vector search.</li>
<li><b>Content generation:</b> GPT-4o + Guardrails AI. Quality output with safety rails.</li>
</ul>
</div>
<div class="rounded-xl border border-slate-200 p-5 bg-slate-50">
<h3 class="font-semibold text-slate-900 mb-3">Pick by constraint</h3>
<ul class="space-y-2 text-sm text-slate-700">
<li><b>Lowest cost:</b> Llama 3 + Ollama + pgvector. Open models, local inference, no API bills.</li>
<li><b>Fastest inference:</b> Groq or vLLM. Custom hardware or optimized serving engine.</li>
<li><b>Best quality:</b> Claude / GPT-4o. Top closed models for reasoning and generation.</li>
<li><b>Privacy-first:</b> Self-hosted Llama + Qdrant. No data leaves your infrastructure.</li>
<li><b>Enterprise compliance:</b> Azure OpenAI + Guardrails. SOC2, data residency, safety rails.</li>
</ul>
</div>
<div class="rounded-xl border border-rose-200 p-5 bg-rose-50">
<h3 class="font-semibold text-rose-900 mb-3">Anti-patterns to avoid</h3>
<ul class="space-y-2 text-sm text-rose-900/90">
<li>Building a custom vector DB instead of using Pinecone/Qdrant/pgvector.</li>
<li>Fine-tuning before trying prompting and RAG first.</li>
<li>Using agents for simple tasks that a single LLM call can handle.</li>
<li>Skipping eval — "vibes-based" quality assessment doesn't scale.</li>
<li>Treating all LLMs as interchangeable — they have different strengths.</li>
</ul>
</div>
</div>
</div>
</section>
<footer class="bg-slate-900 text-slate-400 text-sm">
<div class="max-w-7xl mx-auto px-6 py-8">
<div class="flex items-center justify-between flex-wrap gap-4 mb-4">
<div class="font-semibold text-white">⚡ Open TechStack</div>
<div class="flex items-center gap-4">
<a href="index.html" class="hover:text-white transition">Home</a>
<a href="Frontend-Stacks-Visual-Guide.html" class="hover:text-white transition">Frontend Guide</a>
<a href="Backend-Stacks-Visual-Guide.html" class="hover:text-white transition">Backend Guide</a>
<a href="AIML-Stacks-Visual-Guide.html" class="text-slate-200">AI/ML Guide</a>
</div>
</div>
<div class="flex items-center justify-between flex-wrap gap-3 pt-4 border-t border-slate-800">
<div>Open source · <a href="https://github.com/JagadeepPortfolio/awesome-vibecoding-techstacks" class="text-violet-400 hover:text-violet-300 transition" target="_blank" rel="noopener">Star on GitHub</a> · Contributions welcome</div>
<div class="text-xs">Built with vanilla HTML, Tailwind CSS, and zero dependencies</div>
</div>
</div>
</footer>
<!-- FLOATING SHARE BAR -->
<div id="shareBar" class="fixed bottom-0 left-0 right-0 z-50 bg-slate-900/95 backdrop-blur border-t border-slate-700 transform translate-y-full" style="transition:transform .3s ease">
<div class="max-w-7xl mx-auto px-6 py-3 flex items-center justify-between gap-4">
<div class="text-white text-sm font-medium hidden sm:block">Found this useful? Share it!</div>
<div class="flex items-center gap-2 flex-wrap">
<a id="shareTwitter" target="_blank" rel="noopener" class="inline-flex items-center gap-1.5 px-3 py-1.5 bg-[#1DA1F2] text-white text-xs font-semibold rounded-lg hover:opacity-90 transition"><svg class="w-3.5 h-3.5" fill="currentColor" viewBox="0 0 24 24"><path d="M18.244 2.25h3.308l-7.227 8.26 8.502 11.24H16.17l-5.214-6.817L4.99 21.75H1.68l7.73-8.835L1.254 2.25H8.08l4.713 6.231zm-1.161 17.52h1.833L7.084 4.126H5.117z"/></svg>Tweet</a>
<a id="shareReddit" target="_blank" rel="noopener" class="inline-flex items-center gap-1.5 px-3 py-1.5 bg-[#FF4500] text-white text-xs font-semibold rounded-lg hover:opacity-90 transition">Reddit</a>
<a id="shareLinkedin" target="_blank" rel="noopener" class="inline-flex items-center gap-1.5 px-3 py-1.5 bg-[#0A66C2] text-white text-xs font-semibold rounded-lg hover:opacity-90 transition">LinkedIn</a>
<a href="https://github.com/JagadeepPortfolio/awesome-vibecoding-techstacks" target="_blank" rel="noopener" class="inline-flex items-center gap-1.5 px-3 py-1.5 bg-white text-slate-900 text-xs font-semibold rounded-lg hover:bg-slate-100 transition">Star on GitHub</a>
</div>
</div>
</div>
<script>
(function(){
const bar=document.getElementById("shareBar");const url=encodeURIComponent(location.href);const title=encodeURIComponent(document.title);
const text=encodeURIComponent("Check out this interactive guide to modern tech stacks — 200+ technologies compared with mind maps and filters");
document.getElementById("shareTwitter").href=`https://x.com/intent/tweet?text=${text}&url=${url}`;
document.getElementById("shareReddit").href=`https://reddit.com/submit?url=${url}&title=${title}`;
document.getElementById("shareLinkedin").href=`https://www.linkedin.com/sharing/share-offsite/?url=${url}`;
let shown=false;window.addEventListener("scroll",()=>{if(window.scrollY>400&&!shown){shown=true;bar.style.transform="translateY(0)";}if(window.scrollY<200&&shown){shown=false;bar.style.transform="translateY(100%)";}});
})();
</script>
<script>
/* ================= DATA ================= */
const CATEGORIES = [
{ id:"llm", label:"LLM Providers", cls:"cat-llm" },
{ id:"fw", label:"Frameworks & Orchestration", cls:"cat-fw" },
{ id:"vec", label:"Vector Databases", cls:"cat-vec" },
{ id:"embed", label:"Embeddings & Models", cls:"cat-embed" },
{ id:"rag", label:"RAG & Retrieval", cls:"cat-rag" },
{ id:"agent", label:"Agents & Tools", cls:"cat-agent" },
{ id:"fine", label:"Fine-tuning & Training", cls:"cat-fine" },
{ id:"eval", label:"Eval & Monitoring", cls:"cat-eval" },
{ id:"infra", label:"Inference & Infra", cls:"cat-infra" },
{ id:"data", label:"Data & Labeling", cls:"cat-data" },
];
const STAGES = [
{ id:"hobby", label:"Hobby", cost:"$0", tag:"Prototype", desc:"Free-tier APIs, quick experiments, prompt engineering." },
{ id:"mvp", label:"MVP", cost:"$20–$100", tag:"First users", desc:"Production API keys, basic RAG, first fine-tunes." },
{ id:"growth", label:"Growth", cost:"$100–$2k", tag:"Scaling", desc:"Custom models, vector DB, eval pipelines, guardrails." },
{ id:"scale", label:"Scale", cost:"$2k+", tag:"Enterprise", desc:"Self-hosted models, GPU clusters, compliance, multi-model routing." },
];
const STAGE_STACKS = {
hobby: { llm:"OpenAI / Claude free tier", framework:"LangChain / Vercel AI SDK", vectordb:"Chroma / in-memory", eval:"Manual / vibes", infra:"API calls only", data:"Manual curation" },
mvp: { llm:"Claude / GPT-4o", framework:"LangChain / LlamaIndex", vectordb:"Pinecone / Weaviate", eval:"LangSmith / Braintrust", infra:"Modal / Replicate", data:"Label Studio" },
growth: { llm:"Claude / GPT-4o + open models", framework:"LangGraph / CrewAI", vectordb:"Pinecone / Qdrant / pgvector", eval:"Braintrust + custom", infra:"Modal / Together AI", data:"Scale AI / Labelbox" },
scale: { llm:"Multi-model routing + self-hosted", framework:"Custom orchestration", vectordb:"Qdrant / Milvus self-hosted", eval:"Full pipeline + human review", infra:"vLLM on GPU clusters", data:"Enterprise data pipelines" },
};
const TECH = [
// LLM Providers
{ id:"openai", name:"OpenAI (GPT-4o / o1)", cat:"llm", summary:"The default LLM provider. GPT-4o for speed, o1 for reasoning. Best ecosystem, most integrations.", pros:["Largest ecosystem","Best function calling","Multimodal"], cons:["Expensive at scale","Closed source","Rate limits"], bestFor:"Most AI applications", free:true, oss:false, prod:true, stages:["hobby","mvp","growth","scale"] },
{ id:"anthropic", name:"Anthropic (Claude)", cat:"llm", summary:"Leading on safety and long context (200K tokens). Claude 4 family dominates coding and analysis tasks.", pros:["200K context","Best for code/analysis","Strong safety"], cons:["Smaller ecosystem","API-only","Pricier than open models"], bestFor:"Code generation, analysis, long-form", free:true, oss:false, prod:true, stages:["hobby","mvp","growth","scale"] },
{ id:"google", name:"Google (Gemini)", cat:"llm", summary:"Gemini 2.5 with massive context windows. Native multimodal, strong at structured tasks.", pros:["Huge context","Native multimodal","Competitive pricing"], cons:["Inconsistent quality","Less developer mindshare"], bestFor:"Multimodal apps, Google Cloud teams", free:true, oss:false, prod:true, stages:["hobby","mvp","growth","scale"] },
{ id:"meta", name:"Meta (Llama 3)", cat:"llm", summary:"Best open-weight model family. Run locally, fine-tune, deploy anywhere. No API dependency.", pros:["Open weights","Fine-tunable","No vendor lock-in"], cons:["Self-hosting complexity","Less capable than top closed models"], bestFor:"Self-hosted AI, privacy-first apps", free:true, oss:true, prod:true, stages:["mvp","growth","scale"] },
{ id:"mistral", name:"Mistral", cat:"llm", summary:"European AI lab. Excellent small-to-medium models. Mixtral MoE architecture.", pros:["Great perf/cost ratio","EU data sovereignty","Open models available"], cons:["Smaller ecosystem","Less brand recognition"], bestFor:"Cost-efficient inference, EU compliance", free:true, oss:true, prod:true, stages:["mvp","growth","scale"] },
{ id:"cohere", name:"Cohere", cat:"llm", summary:"Enterprise-focused. Strong embeddings, reranking, and RAG-specific models.", pros:["Best-in-class embeddings","RAG-optimized","Enterprise features"], cons:["Niche focus","Smaller community"], bestFor:"Enterprise RAG and search", free:true, oss:false, prod:true, stages:["growth","scale"] },
// Frameworks & Orchestration
{ id:"langchain", name:"LangChain", cat:"fw", summary:"Most popular LLM framework. Chains, agents, tools, memory. Huge ecosystem.", pros:["Massive ecosystem","Tons of integrations","Active community"], cons:["Abstraction overhead","Breaking changes","Can be over-engineered"], bestFor:"Prototyping, complex LLM pipelines", free:true, oss:true, prod:true, stages:["hobby","mvp","growth","scale"] },
{ id:"llamaindex", name:"LlamaIndex", cat:"fw", summary:"Data framework for LLMs. Best-in-class for connecting LLMs to your data.", pros:["Great data connectors","RAG-focused","Good abstractions"], cons:["Narrower scope than LangChain","Learning curve"], bestFor:"RAG applications, data-heavy AI apps", free:true, oss:true, prod:true, stages:["hobby","mvp","growth","scale"] },
{ id:"vercelai", name:"Vercel AI SDK", cat:"fw", summary:"Streaming-first, React-native AI SDK. Edge-ready, framework-agnostic.", pros:["Great streaming UX","React integration","Edge-ready"], cons:["Vercel-centric","Less orchestration"], bestFor:"Chat UIs, streaming AI apps", free:true, oss:true, prod:true, stages:["hobby","mvp","growth"] },
{ id:"langgraph", name:"LangGraph", cat:"fw", summary:"Stateful multi-actor agent framework built on LangChain. Graph-based workflows.", pros:["Stateful agents","Checkpointing","Human-in-the-loop"], cons:["LangChain dependency","Complex for simple use cases"], bestFor:"Complex agent workflows", free:true, oss:true, prod:true, stages:["mvp","growth","scale"] },
{ id:"haystack", name:"Haystack", cat:"fw", summary:"Modular NLP/LLM framework by deepset. Pipeline-based, production-focused.", pros:["Production-ready pipelines","Modular","Good eval tools"], cons:["Smaller community","Less hype"], bestFor:"Production NLP/RAG pipelines", free:true, oss:true, prod:true, stages:["mvp","growth","scale"] },
{ id:"semantic", name:"Semantic Kernel", cat:"fw", summary:"Microsoft's LLM orchestration. .NET and Python. Deep Azure integration.", pros:["Enterprise-ready","Azure integration","Good planning"], cons:["Microsoft-centric","Smaller community"], bestFor:".NET/Azure shops", free:true, oss:true, prod:true, stages:["growth","scale"] },
// Vector Databases
{ id:"pinecone", name:"Pinecone", cat:"vec", summary:"Managed vector DB. Serverless, scales to billions. The default choice.", pros:["Fully managed","Scales effortlessly","Great DX"], cons:["Vendor lock-in","Can get expensive","Closed source"], bestFor:"Most production vector search", free:true, oss:false, prod:true, stages:["hobby","mvp","growth","scale"] },
{ id:"qdrant", name:"Qdrant", cat:"vec", summary:"High-performance open-source vector DB. Rust-built, self-hostable.", pros:["Open source","Excellent performance","Rich filtering"], cons:["Self-hosting ops","Smaller managed offering"], bestFor:"Self-hosted vector search", free:true, oss:true, prod:true, stages:["mvp","growth","scale"] },
{ id:"weaviate", name:"Weaviate", cat:"vec", summary:"Open-source vector DB with built-in vectorization modules.", pros:["Built-in vectorizers","GraphQL API","Hybrid search"], cons:["Resource-heavy","Complex config"], bestFor:"Hybrid search applications", free:true, oss:true, prod:true, stages:["mvp","growth","scale"] },
{ id:"chroma", name:"Chroma", cat:"vec", summary:"Lightweight, developer-friendly. Great for prototyping and small datasets.", pros:["Dead simple","Great for prototyping","Python-native"], cons:["Not for large scale","Newer"], bestFor:"Prototyping, small RAG apps", free:true, oss:true, prod:true, stages:["hobby","mvp"] },
{ id:"pgvector", name:"pgvector (Postgres)", cat:"vec", summary:"Vector search as a Postgres extension. No new infra needed.", pros:["Use existing Postgres","No new infra","SQL familiar"], cons:["Not optimized for billions","Limited ANN algorithms"], bestFor:"Teams already on Postgres", free:true, oss:true, prod:true, stages:["hobby","mvp","growth"] },
{ id:"milvus", name:"Milvus / Zilliz", cat:"vec", summary:"Distributed vector DB for enterprise scale. Handles billions of vectors.", pros:["Massive scale","Cloud-native","Rich indexes"], cons:["Complex to operate","Heavy for small use"], bestFor:"Enterprise-scale vector search", free:true, oss:true, prod:true, stages:["growth","scale"] },
// Embeddings & Models
{ id:"openaiembed", name:"OpenAI Embeddings", cat:"embed", summary:"text-embedding-3 family. Best balance of quality and cost.", pros:["Easy to use","Good quality","Multiple sizes"], cons:["API dependency","Cost at scale"], bestFor:"Most embedding needs", free:false, oss:false, prod:true, stages:["hobby","mvp","growth","scale"] },
{ id:"cohereenbed", name:"Cohere Embed v3", cat:"embed", summary:"Best-in-class multilingual embeddings with compression.", pros:["Multilingual","Compression","Top benchmarks"], cons:["API dependency","Less ecosystem"], bestFor:"Multilingual search", free:true, oss:false, prod:true, stages:["mvp","growth","scale"] },
{ id:"sentence", name:"Sentence Transformers", cat:"embed", summary:"Open-source embedding models. Run locally, no API costs.", pros:["Free","Local inference","Huge model hub"], cons:["Self-host complexity","GPU needed for speed"], bestFor:"Cost-sensitive, privacy-first", free:true, oss:true, prod:true, stages:["mvp","growth","scale"] },
{ id:"voyage", name:"Voyage AI", cat:"embed", summary:"Specialized embeddings for code, legal, finance domains.", pros:["Domain-specific","High quality","Good compression"], cons:["Niche","Smaller community"], bestFor:"Domain-specific search", free:true, oss:false, prod:true, stages:["mvp","growth"] },
// RAG & Retrieval
{ id:"langchainrag", name:"LangChain RAG", cat:"rag", summary:"Full RAG pipeline: loaders, splitters, retrievers, chains.", pros:["Comprehensive","Many loaders","Flexible"], cons:["Complexity","Debugging hard"], bestFor:"Custom RAG pipelines", free:true, oss:true, prod:true, stages:["hobby","mvp","growth","scale"] },
{ id:"llamaparse", name:"LlamaParse / LlamaCloud", cat:"rag", summary:"Document parsing for RAG. PDFs, tables, images to structured text.", pros:["Best PDF parsing","Table extraction","Cloud + local"], cons:["Paid for volume","Newer"], bestFor:"Document-heavy RAG", free:true, oss:false, prod:true, stages:["mvp","growth","scale"] },
{ id:"unstructured", name:"Unstructured.io", cat:"rag", summary:"Open-source document pre-processing. ETL for LLMs.", pros:["Open source","Many formats","Production-ready"], cons:["Resource-heavy","Config complexity"], bestFor:"Enterprise document ingestion", free:true, oss:true, prod:true, stages:["mvp","growth","scale"] },
{ id:"ragas", name:"RAGAS", cat:"rag", summary:"RAG evaluation framework. Faithfulness, relevancy, context metrics.", pros:["Standard RAG metrics","Easy to use","Growing adoption"], cons:["Evaluation is imperfect","Limited scope"], bestFor:"RAG quality measurement", free:true, oss:true, prod:true, stages:["mvp","growth","scale"] },
// Agents & Tools
{ id:"crewai", name:"CrewAI", cat:"agent", summary:"Multi-agent orchestration. Roles, goals, tasks for agent teams.", pros:["Intuitive role model","Easy multi-agent","Growing fast"], cons:["Newer","Less battle-tested"], bestFor:"Multi-agent systems", free:true, oss:true, prod:true, stages:["mvp","growth"] },
{ id:"autogen", name:"AutoGen (Microsoft)", cat:"agent", summary:"Multi-agent conversation framework. Agents collaborate via chat.", pros:["Flexible patterns","Microsoft backing","Good research tool"], cons:["Complex setup","More research than production"], bestFor:"Research, complex agent patterns", free:true, oss:true, prod:true, stages:["mvp","growth"] },
{ id:"openaiassist", name:"OpenAI Assistants API", cat:"agent", summary:"Managed agent runtime. File search, code interpreter, function calling.", pros:["Fully managed","Built-in tools","Easy to start"], cons:["OpenAI lock-in","Limited customization","Cost"], bestFor:"Quick agent prototypes", free:false, oss:false, prod:true, stages:["hobby","mvp","growth"] },
{ id:"claudemcp", name:"Claude MCP (Model Context Protocol)", cat:"agent", summary:"Open standard for connecting AI to tools and data sources. Universal tool protocol.", pros:["Open standard","Growing adoption","Tool interop"], cons:["Newer","Ecosystem building"], bestFor:"Tool-connected AI apps", free:true, oss:true, prod:true, stages:["hobby","mvp","growth","scale"] },
{ id:"browseruse", name:"Browser Use / Playwright AI", cat:"agent", summary:"AI agents that browse the web. Autonomous web interaction.", pros:["Web automation","Visual understanding","Flexible"], cons:["Unreliable on complex sites","Expensive"], bestFor:"Web scraping, testing, automation", free:true, oss:true, prod:true, stages:["mvp","growth"] },
// Fine-tuning & Training
{ id:"openaifineTune", name:"OpenAI Fine-tuning", cat:"fine", summary:"Fine-tune GPT models via API. No GPU management.", pros:["Simple API","No infra","Good results"], cons:["Expensive","Limited control","Data sent to OpenAI"], bestFor:"Quick fine-tunes without infra", free:false, oss:false, prod:true, stages:["mvp","growth"] },
{ id:"huggingface", name:"Hugging Face", cat:"fine", summary:"The GitHub of ML. Models, datasets, training tools, Spaces.", pros:["Massive model hub","Great community","Full toolchain"], cons:["Fragmented docs","GPU costs"], bestFor:"Any ML/AI project", free:true, oss:true, prod:true, stages:["hobby","mvp","growth","scale"] },
{ id:"axolotl", name:"Axolotl", cat:"fine", summary:"Streamlined fine-tuning tool. YAML config, supports LoRA/QLoRA/full.", pros:["Easy config","Multi-method","Community-driven"], cons:["Smaller project","Breaking changes"], bestFor:"Efficient open-model fine-tuning", free:true, oss:true, prod:true, stages:["mvp","growth","scale"] },
{ id:"unsloth", name:"Unsloth", cat:"fine", summary:"2-5x faster fine-tuning with 80% less memory. LoRA/QLoRA optimized.", pros:["Much faster","Less VRAM","Easy to use"], cons:["Limited model support","Newer"], bestFor:"Fast, cheap fine-tuning", free:true, oss:true, prod:true, stages:["mvp","growth"] },
// Eval & Monitoring
{ id:"langsmith", name:"LangSmith", cat:"eval", summary:"LangChain's observability platform. Traces, evals, datasets, monitoring.", pros:["Deep LangChain integration","Great UI","Comprehensive"], cons:["LangChain-centric","Pricing at scale"], bestFor:"LangChain-based apps", free:true, oss:false, prod:true, stages:["mvp","growth","scale"] },
{ id:"braintrust", name:"Braintrust", cat:"eval", summary:"AI product eval platform. A/B testing, scoring, logging for LLM apps.", pros:["Great eval framework","Good DX","Reasonable pricing"], cons:["Newer","Smaller community"], bestFor:"LLM app evaluation", free:true, oss:false, prod:true, stages:["mvp","growth","scale"] },
{ id:"phoenix", name:"Arize Phoenix", cat:"eval", summary:"Open-source LLM observability. Traces, evals, embeddings visualization.", pros:["Open source","Great visualizations","Traces"], cons:["Self-host complexity","Fewer integrations"], bestFor:"Open-source LLM monitoring", free:true, oss:true, prod:true, stages:["mvp","growth","scale"] },
{ id:"helicone", name:"Helicone", cat:"eval", summary:"LLM proxy for logging, caching, rate limiting. One-line integration.", pros:["Dead simple setup","Caching saves money","Good analytics"], cons:["Proxy adds latency","Newer"], bestFor:"Quick LLM observability", free:true, oss:true, prod:true, stages:["hobby","mvp","growth"] },
{ id:"guardrails", name:"Guardrails AI / NeMo Guardrails", cat:"eval", summary:"Output validation and safety rails for LLM apps.", pros:["Safety-first","Structured output","Open source"], cons:["Adds latency","Config overhead"], bestFor:"Production LLM safety", free:true, oss:true, prod:true, stages:["growth","scale"] },
// Inference & Infra
{ id:"modal", name:"Modal", cat:"infra", summary:"Serverless GPU compute. Deploy Python functions to GPUs in seconds.", pros:["Incredible DX","Pay-per-use GPU","Fast cold starts"], cons:["Vendor lock-in","Pricing at scale"], bestFor:"GPU serverless, model serving", free:true, oss:false, prod:true, stages:["hobby","mvp","growth"] },
{ id:"replicate", name:"Replicate", cat:"infra", summary:"Run open models via API. One-click deploy for any model.", pros:["Easy model deploy","Pay-per-use","Huge model library"], cons:["Cold starts","Less control"], bestFor:"Quick model deployment", free:true, oss:false, prod:true, stages:["hobby","mvp","growth"] },
{ id:"together", name:"Together AI", cat:"infra", summary:"Fast inference for open models. Competitive pricing, good throughput.", pros:["Fast inference","Good pricing","Many models"], cons:["Less flexibility","API-only"], bestFor:"Open model inference at scale", free:true, oss:false, prod:true, stages:["mvp","growth","scale"] },
{ id:"vllm", name:"vLLM", cat:"infra", summary:"High-throughput LLM serving engine. PagedAttention, continuous batching.", pros:["Fastest serving","Open source","Production-proven"], cons:["GPU ops knowledge needed","Self-hosted only"], bestFor:"Self-hosted high-throughput inference", free:true, oss:true, prod:true, stages:["growth","scale"] },
{ id:"ollama", name:"Ollama", cat:"infra", summary:"Run LLMs locally with one command. Mac/Linux/Windows.", pros:["Dead simple","Local/private","Free"], cons:["Limited to local hardware","Not for production scale"], bestFor:"Local development, privacy", free:true, oss:true, prod:true, stages:["hobby","mvp"] },
{ id:"groq", name:"Groq", cat:"infra", summary:"Custom LPU hardware. Fastest inference available.", pros:["Blazing fast","Simple API","Good free tier"], cons:["Limited models","Hardware-specific","Capacity limits"], bestFor:"Latency-critical inference", free:true, oss:false, prod:true, stages:["hobby","mvp","growth"] },
// Data & Labeling
{ id:"labelstudio", name:"Label Studio", cat:"data", summary:"Open-source data labeling. Text, image, audio, video.", pros:["Open source","Multi-modal","Self-hostable"], cons:["UI can be clunky","Scale limits"], bestFor:"In-house labeling", free:true, oss:true, prod:true, stages:["mvp","growth"] },
{ id:"scaleai", name:"Scale AI", cat:"data", summary:"Enterprise data labeling and AI data engine.", pros:["Highest quality","Enterprise scale","RLHF support"], cons:["Expensive","Enterprise sales cycle"], bestFor:"Enterprise training data", free:false, oss:false, prod:true, stages:["growth","scale"] },
{ id:"argilla", name:"Argilla", cat:"data", summary:"Open-source feedback and curation for LLMs. RLHF data collection.", pros:["Open source","RLHF-focused","Hugging Face integration"], cons:["Smaller community","Self-host ops"], bestFor:"LLM feedback loops, RLHF data", free:true, oss:true, prod:true, stages:["mvp","growth","scale"] },
{ id:"snorkel", name:"Snorkel AI", cat:"data", summary:"Programmatic labeling and data-centric AI.", pros:["Programmatic labeling","Enterprise-ready","Reduces manual work"], cons:["Enterprise pricing","Learning curve"], bestFor:"Large-scale data labeling automation", free:false, oss:false, prod:true, stages:["growth","scale"] },
];
/* ================= STATE ================= */
const state = { q:"", cats:new Set(), free:false, oss:false, prod:false, stage:null };
function catOf(id){return CATEGORIES.find(c=>c.id===id);}
function matches(t){
if (state.q){
const q=state.q.toLowerCase();
const hay=(t.name+" "+t.summary+" "+t.bestFor+" "+t.cat).toLowerCase();
if (!hay.includes(q)) return false;
}
if (state.cats.size && !state.cats.has(t.cat)) return false;
if (state.free && !t.free) return false;
if (state.oss && !t.oss) return false;
if (state.prod && !t.prod) return false;
if (state.stage && !t.stages.includes(state.stage)) return false;
return true;
}
/* ================= CHIPS ================= */
const chipsEl = document.getElementById("catChips");
CATEGORIES.forEach(c=>{
const b=document.createElement("button");
b.className=`filter-chip ${c.cls} chip rounded-full px-3 py-1 font-medium`;
b.textContent=c.label; b.dataset.cat=c.id;
b.addEventListener("click",()=>{
if (state.cats.has(c.id)){state.cats.delete(c.id);b.classList.remove("active");}
else {state.cats.add(c.id);b.classList.add("active");}
render();
});
chipsEl.appendChild(b);
});
/* ================= MINDMAP ================= */
const svg=document.getElementById("mindmap");
const NS="http://www.w3.org/2000/svg";
const CX=600, CY=380;
const PRIMARY=[
{cat:"llm", a:-Math.PI/2, r:220},
{cat:"fw", a:-Math.PI/2+(2*Math.PI/5), r:220},
{cat:"vec", a:-Math.PI/2+(4*Math.PI/5), r:220},
{cat:"agent", a:-Math.PI/2+(6*Math.PI/5), r:220},
{cat:"infra", a:-Math.PI/2+(8*Math.PI/5), r:220},
];
const SECONDARY=["embed","rag","fine","eval","data"];
const COLORS={llm:"#6366f1",fw:"#06b6d4",vec:"#10b981",embed:"#8b5cf6",rag:"#f59e0b",agent:"#e11d48",fine:"#ec4899",eval:"#0ea5e9",infra:"#14b8a6",data:"#f97316"};
function buildMindmap(){
svg.innerHTML="";
const defs=document.createElementNS(NS,"defs");
defs.innerHTML=`<radialGradient id="g1"><stop offset="0%" stop-color="#7c3aed"/><stop offset="100%" stop-color="#7c3aed" stop-opacity="0"/></radialGradient>`;
svg.appendChild(defs);
const bg=document.createElementNS(NS,"circle");
bg.setAttribute("cx",CX);bg.setAttribute("cy",CY);bg.setAttribute("r","260");
bg.setAttribute("fill","url(#g1)");bg.setAttribute("opacity","0.15");
svg.appendChild(bg);
// center
const g=document.createElementNS(NS,"g");
const c=document.createElementNS(NS,"circle");
c.setAttribute("cx",CX);c.setAttribute("cy",CY);c.setAttribute("r","58");c.setAttribute("fill","#0f172a");
const t=document.createElementNS(NS,"text");
t.setAttribute("x",CX);t.setAttribute("y",CY-4);t.setAttribute("text-anchor","middle");t.setAttribute("fill","white");t.setAttribute("font-weight","700");t.setAttribute("font-size","18");t.textContent="AI / ML";
const s=document.createElementNS(NS,"text");
s.setAttribute("x",CX);s.setAttribute("y",CY+16);s.setAttribute("text-anchor","middle");s.setAttribute("fill","#94a3b8");s.setAttribute("font-size","11");s.textContent="2026";
g.appendChild(c);g.appendChild(t);g.appendChild(s);svg.appendChild(g);
PRIMARY.forEach(br=>{
const bx=CX+Math.cos(br.a)*br.r, by=CY+Math.sin(br.a)*br.r;
const line=document.createElementNS(NS,"line");
line.setAttribute("x1",CX);line.setAttribute("y1",CY);line.setAttribute("x2",bx);line.setAttribute("y2",by);
line.setAttribute("stroke",COLORS[br.cat]);line.setAttribute("stroke-width","3");line.setAttribute("opacity","0.5");
line.classList.add("link");line.dataset.cat=br.cat;svg.appendChild(line);
const hub=document.createElementNS(NS,"g");
hub.classList.add("node");hub.dataset.cat=br.cat;
const hc=document.createElementNS(NS,"circle");
hc.setAttribute("cx",bx);hc.setAttribute("cy",by);hc.setAttribute("r","38");hc.setAttribute("fill",COLORS[br.cat]);
const ht=document.createElementNS(NS,"text");
ht.setAttribute("x",bx);ht.setAttribute("y",by+4);ht.setAttribute("text-anchor","middle");ht.setAttribute("fill","white");ht.setAttribute("font-weight","700");ht.setAttribute("font-size","11");
ht.textContent=catOf(br.cat).label.split(" ")[0];
hub.appendChild(hc);hub.appendChild(ht);
hub.addEventListener("click",()=>{
if (state.cats.has(br.cat)) state.cats.delete(br.cat); else state.cats.add(br.cat);
document.querySelectorAll(".filter-chip").forEach(el=>{if(el.dataset.cat===br.cat) el.classList.toggle("active",state.cats.has(br.cat));});
render();
});
svg.appendChild(hub);
const items=TECH.filter(x=>x.cat===br.cat);
items.forEach((it,i)=>{
const spread=Math.min(items.length,6);
const off=(i-(spread-1)/2)*0.26;
const la=br.a+off;
const lr=br.r+125;
const lx=CX+Math.cos(la)*lr, ly=CY+Math.sin(la)*lr;
const ll=document.createElementNS(NS,"line");
ll.setAttribute("x1",bx);ll.setAttribute("y1",by);ll.setAttribute("x2",lx);ll.setAttribute("y2",ly);
ll.setAttribute("stroke",COLORS[br.cat]);ll.setAttribute("stroke-width","1.5");ll.setAttribute("opacity","0.4");
ll.classList.add("link");ll.dataset.tech=it.id;ll.dataset.cat=it.cat;svg.appendChild(ll);
const lg=document.createElementNS(NS,"g");
lg.classList.add("node");lg.dataset.tech=it.id;lg.dataset.cat=it.cat;
const w=Math.max(70,Math.min(150,it.name.length*6.3));
const r=document.createElementNS(NS,"rect");
r.setAttribute("x",lx-w/2);r.setAttribute("y",ly-14);r.setAttribute("rx","12");r.setAttribute("ry","12");
r.setAttribute("width",w);r.setAttribute("height","28");
r.setAttribute("fill","white");r.setAttribute("stroke",COLORS[br.cat]);r.setAttribute("stroke-width","2");
const tx=document.createElementNS(NS,"text");
tx.setAttribute("x",lx);tx.setAttribute("y",ly+4);tx.setAttribute("text-anchor","middle");
tx.setAttribute("fill","#0f172a");tx.setAttribute("font-size","10.5");tx.setAttribute("font-weight","600");
tx.textContent=it.name.length>20?it.name.slice(0,19)+"…":it.name;
lg.appendChild(r);lg.appendChild(tx);
lg.addEventListener("click",()=>{
const card=document.getElementById("card-"+it.id);
if (card){card.scrollIntoView({behavior:"smooth",block:"center"});card.classList.remove("flash");void card.offsetWidth;card.classList.add("flash");}
});
svg.appendChild(lg);
});
});
SECONDARY.forEach((cid,i)=>{
const angle=Math.PI*0.58+(i-(SECONDARY.length-1)/2)*0.17;
const r=340;
const x=CX+Math.cos(angle)*r, y=CY+Math.sin(angle)*r;
const line=document.createElementNS(NS,"line");
line.setAttribute("x1",CX);line.setAttribute("y1",CY);line.setAttribute("x2",x);line.setAttribute("y2",y);
line.setAttribute("stroke",COLORS[cid]);line.setAttribute("stroke-width","2");line.setAttribute("opacity","0.35");
line.classList.add("link");line.dataset.cat=cid;svg.appendChild(line);
const g=document.createElementNS(NS,"g");
g.classList.add("node");g.dataset.cat=cid;
const c=document.createElementNS(NS,"circle");
c.setAttribute("cx",x);c.setAttribute("cy",y);c.setAttribute("r","26");c.setAttribute("fill",COLORS[cid]);
const tx=document.createElementNS(NS,"text");
tx.setAttribute("x",x);tx.setAttribute("y",y+4);tx.setAttribute("text-anchor","middle");tx.setAttribute("fill","white");tx.setAttribute("font-weight","700");tx.setAttribute("font-size","10");
tx.textContent=catOf(cid).label.split(" ")[0];
g.appendChild(c);g.appendChild(tx);
g.addEventListener("click",()=>{
if (state.cats.has(cid)) state.cats.delete(cid); else state.cats.add(cid);
document.querySelectorAll(".filter-chip").forEach(el=>{if(el.dataset.cat===cid) el.classList.toggle("active",state.cats.has(cid));});
render();
});
svg.appendChild(g);
});
}
function updateMindmap(){
svg.querySelectorAll(".node").forEach(n=>{
const t=n.dataset.tech?TECH.find(x=>x.id===n.dataset.tech):null;
let visible=true;
if (t) visible=matches(t);
else if (n.dataset.cat) visible=TECH.some(x=>x.cat===n.dataset.cat&&matches(x));
n.classList.toggle("dim",!visible);
});
svg.querySelectorAll(".link").forEach(l=>{
const t=l.dataset.tech?TECH.find(x=>x.id===l.dataset.tech):null;
let visible=true;
if (t) visible=matches(t);
else if (l.dataset.cat) visible=TECH.some(x=>x.cat===l.dataset.cat&&matches(x));
l.classList.toggle("dim",!visible);
});
}
/* ================= STAGES ================= */
const stageTabsEl=document.getElementById("stageTabs");
STAGES.forEach((s,i)=>{
const b=document.createElement("button");
b.className="stage-btn px-4 py-2 rounded-full border border-slate-300 bg-white text-sm font-semibold hover:border-slate-900 transition";
b.innerHTML=`<span class="mr-1">${i+1}</span>${s.label} <span class="text-slate-400 font-normal ml-1">${s.cost}</span>`;
b.dataset.stage=s.id;
b.addEventListener("click",()=>{state.stage=state.stage===s.id?null:s.id;render();});
stageTabsEl.appendChild(b);
});
function renderStagePanel(){
const panel=document.getElementById("stagePanel");
const s=STAGES.find(x=>x.id===state.stage);
if (!s){
panel.innerHTML=`
<div class="text-center py-6">
<div class="text-slate-500 text-sm">Pick a stage above to see the recommended stack and filter cards to what fits.</div>
<div class="mt-4 grid grid-cols-2 md:grid-cols-4 gap-3 max-w-3xl mx-auto">
${STAGES.map(st=>`
<div class="rounded-xl border border-slate-200 bg-white p-3 text-left">
<div class="text-xs text-slate-500">${st.tag}</div>
<div class="text-sm font-bold text-slate-900">${st.label}</div>
<div class="text-xs text-slate-500 mt-1">${st.cost}/mo</div>
</div>`).join("")}
</div>
</div>`;
return;
}
const stack=STAGE_STACKS[s.id];
panel.innerHTML=`
<div class="flex items-start justify-between gap-4 flex-wrap">
<div class="max-w-2xl">
<div class="text-xs uppercase tracking-wider text-violet-600 font-semibold">${s.tag}</div>
<h3 class="text-2xl font-bold text-slate-900 mt-1">${s.label} stage · ${s.cost}/month</h3>
<p class="text-slate-600 mt-2">${s.desc}</p>
</div>
<div class="text-right">
<div class="text-5xl font-extrabold text-slate-900">${s.cost}</div>
<div class="text-xs text-slate-500">typical monthly spend</div>
</div>
</div>
<div class="mt-5 grid grid-cols-2 md:grid-cols-3 lg:grid-cols-6 gap-3">
${Object.entries(stack).map(([k,v])=>`
<div class="rounded-lg border border-slate-200 bg-white p-3">
<div class="text-[10px] uppercase tracking-wide text-slate-500 font-semibold">${k}</div>
<div class="text-sm font-semibold text-slate-900 mt-1">${v}</div>
</div>`).join("")}
</div>`;
}
function updateStageTabs(){
document.querySelectorAll(".stage-btn").forEach(b=>b.classList.toggle("active",b.dataset.stage===state.stage));
}
/* ================= GRID ================= */
const grid=document.getElementById("grid");
function buildGrid(){
grid.innerHTML="";
TECH.forEach(t=>{
const c=catOf(t.cat);
const el=document.createElement("div");
el.id="card-"+t.id;
el.className=`card ${c.cls} bg-white rounded-xl border border-slate-200 p-5 shadow-sm hover:shadow-md`;
el.innerHTML=`
<div class="flex items-start justify-between gap-2 mb-2">
<h3 class="text-base font-bold text-slate-900">${t.name}</h3>
<span class="chip text-[10px] font-semibold uppercase tracking-wide px-2 py-0.5 rounded-full whitespace-nowrap">${c.label}</span>
</div>
<p class="text-sm text-slate-600 leading-relaxed">${t.summary}</p>
<div class="mt-3 grid grid-cols-2 gap-3 text-xs">
<div>
<div class="font-semibold text-emerald-700 mb-1">Pros</div>
<ul class="text-slate-600 space-y-0.5">${t.pros.map(p=>`<li>• ${p}</li>`).join("")}</ul>
</div>
<div>
<div class="font-semibold text-rose-700 mb-1">Cons</div>
<ul class="text-slate-600 space-y-0.5">${t.cons.map(p=>`<li>• ${p}</li>`).join("")}</ul>
</div>
</div>
<div class="mt-3 text-xs text-slate-500"><b class="text-slate-700">Best for:</b> ${t.bestFor}</div>
<div class="mt-3 flex flex-wrap gap-1">
${t.free ? '<span class="text-[10px] bg-emerald-100 text-emerald-700 rounded px-1.5 py-0.5 font-semibold">FREE TIER</span>' : ''}
${t.oss ? '<span class="text-[10px] bg-sky-100 text-sky-700 rounded px-1.5 py-0.5 font-semibold">OPEN SOURCE</span>' : ''}
${t.prod ? '<span class="text-[10px] bg-indigo-100 text-indigo-700 rounded px-1.5 py-0.5 font-semibold">PROD-PROVEN</span>' : ''}
</div>
<div class="mt-2 flex gap-1">
${STAGES.map(s=>{
const ok=t.stages.includes(s.id);
return `<span class="text-[10px] px-1.5 py-0.5 rounded ${ok?'bg-slate-900 text-white':'bg-slate-100 text-slate-400'}">${s.label}</span>`;
}).join("")}
</div>`;
grid.appendChild(el);
});
}
function updateGrid(){
let shown=0;
TECH.forEach(t=>{
const el=document.getElementById("card-"+t.id);
if (!el) return;
const ok=matches(t);
el.classList.toggle("hidden-card",!ok);
if (ok) shown++;
});
document.getElementById("gridCount").textContent=`${shown} of ${TECH.length} shown`;
document.getElementById("countLabel").textContent=`${shown} of ${TECH.length} technologies`;
}
/* ================= RENDER ================= */
function render(){
updateStageTabs();
renderStagePanel();
updateMindmap();
updateGrid();
const p=[];
if (state.stage) p.push("stage="+state.stage);
history.replaceState(null,"",p.length?"#"+p.join("&"):"#");
}
/* ================= EVENTS ================= */
document.getElementById("search").addEventListener("input",e=>{state.q=e.target.value;render();});
document.getElementById("f-free").addEventListener("change",e=>{state.free=e.target.checked;render();});
document.getElementById("f-oss").addEventListener("change",e=>{state.oss=e.target.checked;render();});
document.getElementById("f-prod").addEventListener("change",e=>{state.prod=e.target.checked;render();});
document.getElementById("reset").addEventListener("click",()=>{
state.q="";state.cats.clear();state.free=state.oss=state.prod=false;state.stage=null;
document.getElementById("search").value="";
["f-free","f-oss","f-prod"].forEach(id=>document.getElementById(id).checked=false);
document.querySelectorAll(".filter-chip").forEach(el=>el.classList.remove("active"));
render();
});
document.addEventListener("keydown",e=>{
if (e.key==="/" && document.activeElement.tagName!=="INPUT"){e.preventDefault();document.getElementById("search").focus();}
if (e.key==="Escape") document.getElementById("reset").click();
if (["1","2","3","4"].includes(e.key) && document.activeElement.tagName!=="INPUT"){
state.stage=STAGES[parseInt(e.key,10)-1].id;render();
}
});
(function(){const h=location.hash.slice(1);if (h) h.split("&").forEach(p=>{const[k,v]=p.split("=");if(k==="stage"&&STAGES.find(s=>s.id===v)) state.stage=v;});})();
buildMindmap();
buildGrid();
render();
</script>
</body>
</html>