Tech-tag co-occurrence

Every AI role gets tagged from a 35-term controlled vocabulary spanning agent / eval / training / inference / modality clusters. Tags that frequently appear together in the same JD pull each other close; thicker edges = more co-occurrences. The clusters that emerge organically are the real sub-disciplines of applied AI engineering right now.

All sectors · 4742Enterprise · 1275AI Frontier · 868Industrial · 650Data AI · 526Consumer · 337Banking · 239Fintech · 220Robotics · 127Defense · 115Pharma · 101Media · 73Retail · 62Hospitality · 62Telecom · 27Seattle · 27Aerospace · 14Insurance · 12Healthtech · 7

Showing 650 tagged AI roles in Industrial. Layout is a Fruchterman-Reingold force simulation, run server-side to convergence.

agent_orchestration ↔ rag (75 JDs)agent_orchestration ↔ vector_db (24 JDs)agent_orchestration ↔ llm_observability (83 JDs)agent_orchestration ↔ model_serving (127 JDs)rag ↔ vector_db (28 JDs)llm_observability ↔ rag (47 JDs)model_serving ↔ rag (47 JDs)llm_observability ↔ vector_db (16 JDs)model_serving ↔ vector_db (18 JDs)llm_observability ↔ model_serving (72 JDs)agent_orchestration ↔ multimodal (43 JDs)multimodal ↔ rag (14 JDs)multimodal ↔ vector_db (4 JDs)llm_observability ↔ multimodal (18 JDs)model_serving ↔ multimodal (86 JDs)agent_orchestration ↔ tool_use (61 JDs)agent_orchestration ↔ inference_infra (85 JDs)inference_infra ↔ tool_use (23 JDs)model_serving ↔ tool_use (32 JDs)inference_infra ↔ model_serving (448 JDs)agent_orchestration ↔ agent_research (15 JDs)agent_orchestration ↔ evals (48 JDs)agent_orchestration ↔ embodied_ai (13 JDs)agent_research ↔ tool_use (5 JDs)agent_research ↔ evals (7 JDs)agent_research ↔ llm_observability (9 JDs)agent_research ↔ rag (5 JDs)agent_research ↔ multimodal (4 JDs)agent_research ↔ embodied_ai (2 JDs)evals ↔ tool_use (22 JDs)llm_observability ↔ tool_use (33 JDs)rag ↔ tool_use (34 JDs)multimodal ↔ tool_use (9 JDs)embodied_ai ↔ tool_use (4 JDs)evals ↔ llm_observability (34 JDs)evals ↔ rag (22 JDs)evals ↔ multimodal (28 JDs)embodied_ai ↔ evals (4 JDs)embodied_ai ↔ llm_observability (2 JDs)embodied_ai ↔ multimodal (14 JDs)fine_tuning ↔ model_serving (89 JDs)model_serving ↔ rl_post_training (12 JDs)fine_tuning ↔ inference_infra (63 JDs)inference_infra ↔ rl_post_training (12 JDs)inference_infra ↔ multimodal (64 JDs)fine_tuning ↔ rl_post_training (13 JDs)agent_orchestration ↔ fine_tuning (43 JDs)fine_tuning ↔ multimodal (44 JDs)agent_orchestration ↔ rl_post_training (4 JDs)multimodal ↔ rl_post_training (8 JDs)inference_infra ↔ pretraining (11 JDs)model_serving ↔ pretraining (12 JDs)agent_orchestration ↔ pretraining (2 JDs)fine_tuning ↔ pretraining (15 JDs)embodied_ai ↔ rl_robotics (15 JDs)embodied_ai ↔ vision (7 JDs)embodied_ai ↔ fine_tuning (10 JDs)agent_orchestration ↔ rl_robotics (4 JDs)agent_orchestration ↔ vision (18 JDs)rl_robotics ↔ vision (3 JDs)multimodal ↔ rl_robotics (7 JDs)fine_tuning ↔ rl_robotics (7 JDs)evals ↔ rl_robotics (2 JDs)multimodal ↔ vision (43 JDs)fine_tuning ↔ vision (21 JDs)evals ↔ vision (15 JDs)evals ↔ fine_tuning (31 JDs)fine_tuning ↔ rag (25 JDs)fine_tuning ↔ vector_db (4 JDs)fine_tuning ↔ llm_observability (30 JDs)inference_infra ↔ quantization (27 JDs)evals ↔ inference_infra (41 JDs)model_serving ↔ quantization (27 JDs)evals ↔ model_serving (58 JDs)evals ↔ quantization (2 JDs)inference_infra ↔ llm_observability (45 JDs)llm_observability ↔ vision (9 JDs)model_serving ↔ vision (49 JDs)inference_infra ↔ vision (40 JDs)inference_infra ↔ training_infra (4 JDs)model_serving ↔ training_infra (4 JDs)model_serving ↔ synthetic_data (15 JDs)inference_infra ↔ synthetic_data (12 JDs)embodied_ai ↔ synthetic_data (6 JDs)embodied_ai ↔ model_serving (19 JDs)embodied_ai ↔ inference_infra (14 JDs)agent_orchestration ↔ code_gen (14 JDs)agent_orchestration ↔ synthetic_data (7 JDs)fine_tuning ↔ tool_use (9 JDs)code_gen ↔ tool_use (5 JDs)synthetic_data ↔ tool_use (2 JDs)code_gen ↔ evals (4 JDs)evals ↔ synthetic_data (8 JDs)code_gen ↔ fine_tuning (6 JDs)fine_tuning ↔ synthetic_data (4 JDs)code_gen ↔ rag (7 JDs)rag ↔ synthetic_data (3 JDs)frontier_research ↔ multimodal (14 JDs)fine_tuning ↔ frontier_research (13 JDs)frontier_research ↔ model_serving (8 JDs)model_serving ↔ rl_robotics (9 JDs)model_serving ↔ recommender_systems (12 JDs)inference_infra ↔ recommender_systems (12 JDs)agent_orchestration ↔ recommender_systems (2 JDs)synthetic_data ↔ vision (6 JDs)multimodal ↔ synthetic_data (11 JDs)tool_use ↔ vision (3 JDs)inference_infra ↔ rl_robotics (7 JDs)agent_orchestration ↔ guardrails (17 JDs)inference_infra ↔ rag (27 JDs)guardrails ↔ rag (7 JDs)inference_infra ↔ vector_db (10 JDs)evals ↔ vector_db (3 JDs)guardrails ↔ vector_db (3 JDs)tool_use ↔ vector_db (9 JDs)guardrails ↔ model_serving (15 JDs)guardrails ↔ inference_infra (12 JDs)evals ↔ guardrails (15 JDs)guardrails ↔ llm_observability (15 JDs)guardrails ↔ tool_use (9 JDs)code_gen ↔ vector_db (4 JDs)code_gen ↔ llm_observability (4 JDs)fine_tuning ↔ recommender_systems (2 JDs)agent_research ↔ fine_tuning (6 JDs)agent_research ↔ inference_infra (7 JDs)agent_research ↔ model_serving (12 JDs)agent_research ↔ guardrails (6 JDs)agent_research ↔ code_gen (2 JDs)fine_tuning ↔ guardrails (6 JDs)code_gen ↔ inference_infra (5 JDs)code_gen ↔ model_serving (7 JDs)guardrails ↔ multimodal (4 JDs)code_gen ↔ multimodal (7 JDs)audio_speech ↔ fine_tuning (9 JDs)audio_speech ↔ frontier_research (3 JDs)audio_speech ↔ pretraining (2 JDs)frontier_research ↔ pretraining (15 JDs)fine_tuning ↔ quantization (8 JDs)distillation ↔ fine_tuning (2 JDs)evals ↔ rl_post_training (6 JDs)pretraining ↔ rl_post_training (2 JDs)quantization ↔ rl_post_training (2 JDs)distillation ↔ inference_infra (5 JDs)distillation ↔ model_serving (5 JDs)pretraining ↔ quantization (2 JDs)distillation ↔ quantization (5 JDs)frontier_research ↔ vision (9 JDs)pretraining ↔ vision (6 JDs)multimodal ↔ quantization (8 JDs)guardrails ↔ rl_post_training (2 JDs)llm_observability ↔ rl_post_training (3 JDs)audio_speech ↔ inference_infra (10 JDs)audio_speech ↔ model_serving (11 JDs)quantization ↔ vision (5 JDs)audio_speech ↔ quantization (3 JDs)audio_speech ↔ vision (7 JDs)multimodal ↔ pretraining (7 JDs)rl_post_training ↔ synthetic_data (2 JDs)agent_orchestration ↔ audio_speech (5 JDs)rag ↔ vision (3 JDs)audio_speech ↔ rag (3 JDs)audio_speech ↔ multimodal (7 JDs)audio_speech ↔ evals (3 JDs)audio_speech ↔ llm_observability (2 JDs)llm_observability ↔ synthetic_data (3 JDs)llm_observability ↔ pretraining (2 JDs)agent_research ↔ rl_post_training (2 JDs)frontier_research ↔ inference_infra (6 JDs)frontier_research ↔ llm_observability (4 JDs)agent_orchestration ↔ frontier_research (3 JDs)code_gen ↔ frontier_research (4 JDs)embodied_ai ↔ quantization (2 JDs)distillation ↔ multimodal (2 JDs)code_gen ↔ vision (2 JDs)audio_speech ↔ code_gen (2 JDs)frontier_research ↔ rl_post_training (2 JDs)forecasting ↔ recommender_systems (2 JDs)Agent orchestration N=193 JDs Top co-occur: Model serving ×127 · Inference infra ×85 · LLM observability ×83Agent orchestrationRAG N=80 JDs Top co-occur: Agent orchestration ×75 · LLM observability ×47 · Model serving ×47Vector DB N=29 JDs Top co-occur: RAG ×28 · Agent orchestration ×24 · Model serving ×18LLM observability N=117 JDs Top co-occur: Agent orchestration ×83 · Model serving ×72 · RAG ×47Model serving N=512 JDs Top co-occur: Inference infra ×448 · Agent orchestration ×127 · Fine-tuning ×89Model servingMultimodal N=109 JDs Top co-occur: Model serving ×86 · Inference infra ×64 · Fine-tuning ×44Tool use N=61 JDs Top co-occur: Agent orchestration ×61 · RAG ×34 · LLM observability ×33Inference infra N=450 JDs Top co-occur: Model serving ×448 · Agent orchestration ×85 · Multimodal ×64Inference infraAgent research N=18 JDs Top co-occur: Agent orchestration ×15 · Model serving ×12 · LLM observability ×9Evals N=87 JDs Top co-occur: Model serving ×58 · Agent orchestration ×48 · Inference infra ×41Embodied AI N=38 JDs Top co-occur: Model serving ×19 · RL robotics ×15 · Multimodal ×14Fine-tuning N=112 JDs Top co-occur: Model serving ×89 · Inference infra ×63 · Multimodal ×44RL post-training N=16 JDs Top co-occur: Fine-tuning ×13 · Model serving ×12 · Inference infra ×12Pretraining N=26 JDs Top co-occur: Fine-tuning ×15 · Frontier research ×15 · Model serving ×12RL robotics N=16 JDs Top co-occur: Embodied AI ×15 · Model serving ×9 · Multimodal ×7Vision N=66 JDs Top co-occur: Model serving ×49 · Multimodal ×43 · Inference infra ×40Synthetic data N=25 JDs Top co-occur: Model serving ×15 · Inference infra ×12 · Multimodal ×11Quantization N=28 JDs Top co-occur: Inference infra ×27 · Model serving ×27 · Fine-tuning ×8Training infra N=4 JDs Top co-occur: Inference infra ×4 · Model serving ×4Code gen N=16 JDs Top co-occur: Agent orchestration ×14 · RAG ×7 · Model serving ×7Frontier research N=28 JDs Top co-occur: Pretraining ×15 · Multimodal ×14 · Fine-tuning ×13Recommender systems N=15 JDs Top co-occur: Model serving ×12 · Inference infra ×12 · Agent orchestration ×2Guardrails N=20 JDs Top co-occur: Agent orchestration ×17 · Model serving ×15 · Evals ×15Audio & speech N=15 JDs Top co-occur: Model serving ×11 · Inference infra ×10 · Fine-tuning ×9Distillation N=5 JDs Top co-occur: Inference infra ×5 · Model serving ×5 · Quantization ×5Multi-agent N=2 JDsRLHF N=1 JDsSim-to-real N=1 JDsInterpretability N=2 JDsGPU kernels N=1 JDsSearch & ranking N=1 JDsForecasting N=2 JDs Top co-occur: Recommender systems ×2
32 tags · 177 co-occurrence edges · min edge weight 2. Bubble area ∝ JDs containing tag · edge thickness ∝ co-occurrence count. Hover any node for top-3 partners; click to see the JDs.

Tag velocity · last 4 weeks vs prior 4

Which technologies are hot, which are cooling. Sparkline = 12 weeks of unique roles tagged with each term, last bar on the right is this week. Sorted by absolute pickup. Tags with under 10 lifetime mentions are hidden as noise.

Tag12-week trendLast 4wPrior 4wΔTotal
model_serving
15 roles
Feb 23
12 roles
Mar 2
15 roles
9
14 roles
16
9 roles
23
97 roles
30
50 roles
Apr 6
61 roles
13
58 roles
20
77 roles
27
71 roles
May 4
0 roles
11
254180↑+74512
inference_infra
14 roles
Feb 23
11 roles
Mar 2
11 roles
9
10 roles
16
7 roles
23
87 roles
30
46 roles
Apr 6
55 roles
13
53 roles
20
70 roles
27
60 roles
May 4
0 roles
11
227158↑+69450
agent_orchestration
2 roles
Feb 23
1 role
Mar 2
7 roles
9
6 roles
16
10 roles
23
27 roles
30
19 roles
Apr 6
22 roles
13
24 roles
20
28 roles
27
38 roles
May 4
0 roles
11
10865↑+43193
llm_observability
1 role
Feb 23
1 role
Mar 2
3 roles
9
1 role
16
4 roles
23
17 roles
30
15 roles
Apr 6
13 roles
13
8 roles
20
17 roles
27
24 roles
May 4
0 roles
11
5939↑+20117
rag
1 role
Feb 23
0 roles
Mar 2
2 roles
9
1 role
16
3 roles
23
10 roles
30
12 roles
Apr 6
9 roles
13
8 roles
20
11 roles
27
18 roles
May 4
0 roles
11
4526↑+1980
fine_tuning
4 roles
Feb 23
3 roles
Mar 2
2 roles
9
5 roles
16
4 roles
23
24 roles
30
9 roles
Apr 6
8 roles
13
12 roles
20
18 roles
27
18 roles
May 4
0 roles
11
5641↑+15112
tool_use
1 role
Feb 23
0 roles
Mar 2
0 roles
9
2 roles
16
2 roles
23
10 roles
30
7 roles
Apr 6
9 roles
13
9 roles
20
8 roles
27
10 roles
May 4
0 roles
11
3522↑+1361
evals
1 role
Feb 23
0 roles
Mar 2
3 roles
9
3 roles
16
5 roles
23
11 roles
30
10 roles
Apr 6
12 roles
13
9 roles
20
11 roles
27
17 roles
May 4
0 roles
11
4434↑+1087
embodied_ai
3 roles
Feb 23
1 role
Mar 2
3 roles
9
1 role
16
0 roles
23
6 roles
30
3 roles
Apr 6
3 roles
13
5 roles
20
9 roles
27
3 roles
May 4
0 roles
11
2010↑+1038
vector_db
2 roles
Feb 23
0 roles
Mar 2
2 roles
9
0 roles
16
0 roles
23
4 roles
30
1 role
Apr 6
2 roles
13
1 role
20
3 roles
27
10 roles
May 4
0 roles
11
156↑+929
synthetic_data
1 role
Feb 23
1 role
Mar 2
3 roles
9
0 roles
16
0 roles
23
0 roles
30
5 roles
Apr 6
0 roles
13
4 roles
20
6 roles
27
3 roles
May 4
0 roles
11
135↑+825
agent_research
0 roles
Feb 23
0 roles
Mar 2
0 roles
9
0 roles
16
1 role
23
1 role
30
2 roles
Apr 6
2 roles
13
4 roles
20
4 roles
27
4 roles
May 4
0 roles
11
135↑+818
vision
0 roles
Feb 23
0 roles
Mar 2
3 roles
9
1 role
16
0 roles
23
19 roles
30
6 roles
Apr 6
7 roles
13
5 roles
20
10 roles
27
10 roles
May 4
0 roles
11
3126↑+566
frontier_research
0 roles
Feb 23
0 roles
Mar 2
0 roles
9
0 roles
16
0 roles
23
10 roles
30
1 role
Apr 6
1 role
13
3 roles
20
5 roles
27
7 roles
May 4
0 roles
11
1611↑+528
rl_robotics
3 roles
Feb 23
2 roles
Mar 2
0 roles
9
0 roles
16
0 roles
23
2 roles
30
1 role
Apr 6
1 role
13
1 role
20
3 roles
27
2 roles
May 4
0 roles
11
73↑+416
multimodal
1 role
Feb 23
2 roles
Mar 2
5 roles
9
4 roles
16
3 roles
23
26 roles
30
11 roles
Apr 6
10 roles
13
8 roles
20
13 roles
27
16 roles
May 4
0 roles
11
4644↑+2109
code_gen
0 roles
Feb 23
0 roles
Mar 2
0 roles
9
0 roles
16
1 role
23
3 roles
30
1 role
Apr 6
4 roles
13
2 roles
20
3 roles
27
0 roles
May 4
0 roles
11
86↑+216
recommender_systems
2 roles
Feb 23
0 roles
Mar 2
1 role
9
2 roles
16
0 roles
23
1 role
30
1 role
Apr 6
4 roles
13
0 roles
20
1 role
27
1 role
May 4
0 roles
11
64↑+215
rl_post_training
1 role
Feb 23
0 roles
Mar 2
0 roles
9
0 roles
16
1 role
23
4 roles
30
3 roles
Apr 6
0 roles
13
3 roles
20
1 role
27
3 roles
May 4
0 roles
11
78↓-116
audio_speech
0 roles
Feb 23
3 roles
Mar 2
0 roles
9
0 roles
16
0 roles
23
5 roles
30
2 roles
Apr 6
0 roles
13
3 roles
20
1 role
27
1 role
May 4
0 roles
11
57↓-215
guardrails
1 role
Feb 23
0 roles
Mar 2
0 roles
9
0 roles
16
2 roles
23
4 roles
30
4 roles
Apr 6
2 roles
13
1 role
20
3 roles
27
3 roles
May 4
0 roles
11
811↓-320
pretraining
0 roles
Feb 23
0 roles
Mar 2
0 roles
9
0 roles
16
0 roles
23
12 roles
30
3 roles
Apr 6
1 role
13
2 roles
20
2 roles
27
5 roles
May 4
0 roles
11
1015↓-526
quantization
0 roles
Feb 23
0 roles
Mar 2
1 role
9
0 roles
16
0 roles
23
14 roles
30
2 roles
Apr 6
0 roles
13
5 roles
20
4 roles
27
1 role
May 4
0 roles
11
1016↓-628