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 239 tagged AI roles in Banking. Layout is a Fruchterman-Reingold force simulation, run server-side to convergence.

fine_tuning ↔ inference_infra (63 JDs)fine_tuning ↔ model_serving (93 JDs)fine_tuning ↔ guardrails (59 JDs)fine_tuning ↔ vector_db (104 JDs)fine_tuning ↔ llm_observability (75 JDs)evals ↔ fine_tuning (38 JDs)inference_infra ↔ model_serving (106 JDs)guardrails ↔ inference_infra (64 JDs)inference_infra ↔ vector_db (69 JDs)inference_infra ↔ llm_observability (63 JDs)evals ↔ inference_infra (31 JDs)guardrails ↔ model_serving (73 JDs)model_serving ↔ vector_db (92 JDs)llm_observability ↔ model_serving (81 JDs)evals ↔ model_serving (37 JDs)guardrails ↔ vector_db (70 JDs)guardrails ↔ llm_observability (73 JDs)evals ↔ guardrails (40 JDs)llm_observability ↔ vector_db (81 JDs)evals ↔ vector_db (37 JDs)evals ↔ llm_observability (40 JDs)fine_tuning ↔ rag (51 JDs)rag ↔ vector_db (58 JDs)model_serving ↔ rag (54 JDs)llm_observability ↔ rag (53 JDs)agent_orchestration ↔ model_serving (45 JDs)agent_orchestration ↔ inference_infra (29 JDs)inference_infra ↔ rag (39 JDs)guardrails ↔ rag (39 JDs)agent_orchestration ↔ llm_observability (51 JDs)agent_orchestration ↔ agent_research (11 JDs)agent_research ↔ llm_observability (4 JDs)agent_orchestration ↔ rag (39 JDs)agent_orchestration ↔ fine_tuning (35 JDs)evals ↔ rag (25 JDs)agent_orchestration ↔ evals (20 JDs)agent_orchestration ↔ guardrails (39 JDs)llm_observability ↔ vision (2 JDs)llm_observability ↔ multimodal (5 JDs)rag ↔ vision (2 JDs)multimodal ↔ rag (3 JDs)multimodal ↔ vector_db (4 JDs)fine_tuning ↔ vision (3 JDs)fine_tuning ↔ multimodal (9 JDs)model_serving ↔ vision (3 JDs)model_serving ↔ multimodal (7 JDs)multimodal ↔ vision (3 JDs)agent_orchestration ↔ vector_db (35 JDs)agent_research ↔ rag (3 JDs)agent_research ↔ vector_db (3 JDs)fine_tuning ↔ pretraining (24 JDs)pretraining ↔ rlhf (4 JDs)pretraining ↔ vector_db (22 JDs)fine_tuning ↔ rlhf (6 JDs)rlhf ↔ vector_db (6 JDs)fine_tuning ↔ frontier_research (27 JDs)frontier_research ↔ vector_db (27 JDs)fine_tuning ↔ rl_post_training (16 JDs)fine_tuning ↔ interpretability (6 JDs)fine_tuning ↔ recommender_systems (7 JDs)interpretability ↔ rl_post_training (4 JDs)rl_post_training ↔ vector_db (16 JDs)model_serving ↔ rl_post_training (7 JDs)recommender_systems ↔ rl_post_training (3 JDs)interpretability ↔ vector_db (6 JDs)interpretability ↔ model_serving (3 JDs)interpretability ↔ recommender_systems (3 JDs)recommender_systems ↔ vector_db (6 JDs)model_serving ↔ recommender_systems (6 JDs)frontier_research ↔ pretraining (20 JDs)frontier_research ↔ rlhf (5 JDs)frontier_research ↔ rl_post_training (13 JDs)pretraining ↔ rl_post_training (9 JDs)agent_research ↔ pretraining (2 JDs)agent_research ↔ fine_tuning (2 JDs)agent_research ↔ frontier_research (2 JDs)interpretability ↔ pretraining (2 JDs)frontier_research ↔ interpretability (3 JDs)interpretability ↔ rlhf (2 JDs)frontier_research ↔ inference_infra (2 JDs)frontier_research ↔ model_serving (8 JDs)inference_infra ↔ pretraining (4 JDs)model_serving ↔ pretraining (8 JDs)agent_orchestration ↔ tool_use (8 JDs)rag ↔ tool_use (4 JDs)guardrails ↔ tool_use (2 JDs)llm_observability ↔ tool_use (4 JDs)model_serving ↔ tool_use (4 JDs)tool_use ↔ vector_db (2 JDs)agent_research ↔ evals (2 JDs)recommender_systems ↔ search_ranking (3 JDs)inference_infra ↔ recommender_systems (2 JDs)guardrails ↔ recommender_systems (2 JDs)evals ↔ recommender_systems (2 JDs)agent_research ↔ model_serving (4 JDs)agent_research ↔ inference_infra (2 JDs)agent_orchestration ↔ pretraining (3 JDs)frontier_research ↔ recommender_systems (2 JDs)agent_orchestration ↔ multimodal (2 JDs)fine_tuning ↔ tool_use (3 JDs)agent_orchestration ↔ code_gen (3 JDs)code_gen ↔ llm_observability (3 JDs)Fine-tuning N=127 JDs Top co-occur: Vector DB ×104 · Model serving ×93 · LLM observability ×75Fine-tuningInference infra N=106 JDs Top co-occur: Model serving ×106 · Vector DB ×69 · Guardrails ×64Inference infraModel serving N=152 JDs Top co-occur: Inference infra ×106 · Fine-tuning ×93 · Vector DB ×92Model servingGuardrails N=85 JDs Top co-occur: Model serving ×73 · LLM observability ×73 · Vector DB ×70GuardrailsVector DB N=126 JDs Top co-occur: Fine-tuning ×104 · Model serving ×92 · LLM observability ×81Vector DBLLM observability N=107 JDs Top co-occur: Model serving ×81 · Vector DB ×81 · Fine-tuning ×75LLM observabilityEvals N=49 JDs Top co-occur: Guardrails ×40 · LLM observability ×40 · Fine-tuning ×38EvalsMultimodal N=11 JDs Top co-occur: Fine-tuning ×9 · Model serving ×7 · LLM observability ×5RAG N=72 JDs Top co-occur: Vector DB ×58 · Model serving ×54 · LLM observability ×53RAGAgent orchestration N=87 JDs Top co-occur: LLM observability ×51 · Model serving ×45 · RAG ×39Agent orchestrationAgent research N=12 JDs Top co-occur: Agent orchestration ×11 · LLM observability ×4 · Model serving ×4Audio & speech N=1 JDsRecommender systems N=15 JDs Top co-occur: Fine-tuning ×7 · Vector DB ×6 · Model serving ×6Vision N=3 JDs Top co-occur: Fine-tuning ×3 · Model serving ×3 · Multimodal ×3Pretraining N=24 JDs Top co-occur: Fine-tuning ×24 · Vector DB ×22 · Frontier research ×20RLHF N=6 JDs Top co-occur: Fine-tuning ×6 · Vector DB ×6 · Frontier research ×5Frontier research N=28 JDs Top co-occur: Fine-tuning ×27 · Vector DB ×27 · Pretraining ×20RL post-training N=16 JDs Top co-occur: Fine-tuning ×16 · Vector DB ×16 · Frontier research ×13Interpretability N=6 JDs Top co-occur: Fine-tuning ×6 · Vector DB ×6 · RL post-training ×4Tool use N=8 JDs Top co-occur: Agent orchestration ×8 · RAG ×4 · LLM observability ×4Search & ranking N=4 JDs Top co-occur: Recommender systems ×3Synthetic data N=1 JDsForecasting N=1 JDsMulti-agent N=1 JDsQuantization N=1 JDsCode gen N=3 JDs Top co-occur: Agent orchestration ×3 · LLM observability ×3
26 tags · 102 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
7 roles
Feb 23
4 roles
Mar 2
2 roles
9
7 roles
16
5 roles
23
11 roles
30
20 roles
Apr 6
6 roles
13
13 roles
20
21 roles
27
19 roles
May 4
0 roles
11
5943↑+16152
agent_orchestration
4 roles
Feb 23
1 role
Mar 2
4 roles
9
3 roles
16
2 roles
23
3 roles
30
15 roles
Apr 6
4 roles
13
3 roles
20
21 roles
27
7 roles
May 4
0 roles
11
3424↑+1087
inference_infra
4 roles
Feb 23
3 roles
Mar 2
0 roles
9
7 roles
16
2 roles
23
10 roles
30
11 roles
Apr 6
4 roles
13
7 roles
20
10 roles
27
17 roles
May 4
0 roles
11
3830↑+8106
pretraining
7 roles
Feb 23
4 roles
Mar 2
1 role
9
3 roles
16
0 roles
23
0 roles
30
0 roles
Apr 6
1 role
13
2 roles
20
4 roles
27
1 role
May 4
0 roles
11
83↑+524
fine_tuning
12 roles
Feb 23
5 roles
Mar 2
4 roles
9
8 roles
16
4 roles
23
10 roles
30
12 roles
Apr 6
3 roles
13
14 roles
20
12 roles
27
9 roles
May 4
0 roles
11
3834↑+4127
rag
3 roles
Feb 23
0 roles
Mar 2
1 role
9
3 roles
16
1 role
23
9 roles
30
8 roles
Apr 6
0 roles
13
8 roles
20
10 roles
27
7 roles
May 4
0 roles
11
2521↑+472
agent_research
1 role
Feb 23
0 roles
Mar 2
0 roles
9
0 roles
16
0 roles
23
1 role
30
1 role
Apr 6
1 role
13
0 roles
20
3 roles
27
1 role
May 4
0 roles
11
52↑+312
multimodal
0 roles
Feb 23
0 roles
Mar 2
1 role
9
0 roles
16
0 roles
23
1 role
30
1 role
Apr 6
0 roles
13
0 roles
20
2 roles
27
3 roles
May 4
0 roles
11
52↑+311
frontier_research
6 roles
Feb 23
4 roles
Mar 2
1 role
9
3 roles
16
1 role
23
2 roles
30
1 role
Apr 6
1 role
13
5 roles
20
3 roles
27
0 roles
May 4
0 roles
11
97↑+228
llm_observability
4 roles
Feb 23
2 roles
Mar 2
0 roles
9
6 roles
16
3 roles
23
10 roles
30
15 roles
Apr 6
5 roles
13
8 roles
20
13 roles
27
9 roles
May 4
0 roles
11
3534↑+1107
evals
3 roles
Feb 23
1 role
Mar 2
1 role
9
2 roles
16
2 roles
23
3 roles
30
7 roles
Apr 6
1 role
13
5 roles
20
6 roles
27
2 roles
May 4
0 roles
11
1414·049
rl_post_training
4 roles
Feb 23
0 roles
Mar 2
0 roles
9
2 roles
16
1 role
23
2 roles
30
1 role
Apr 6
1 role
13
2 roles
20
3 roles
27
0 roles
May 4
0 roles
11
66·016
recommender_systems
2 roles
Feb 23
0 roles
Mar 2
1 role
9
1 role
16
1 role
23
1 role
30
3 roles
Apr 6
1 role
13
1 role
20
2 roles
27
1 role
May 4
0 roles
11
56↓-115
vector_db
15 roles
Feb 23
5 roles
Mar 2
3 roles
9
8 roles
16
4 roles
23
13 roles
30
14 roles
Apr 6
2 roles
13
13 roles
20
9 roles
27
9 roles
May 4
0 roles
11
3339↓-6126
guardrails
5 roles
Feb 23
1 role
Mar 2
1 role
9
6 roles
16
3 roles
23
9 roles
30
15 roles
Apr 6
2 roles
13
6 roles
20
6 roles
27
6 roles
May 4
0 roles
11
2033↓-1385