Operation Log
Format: [YYYY-MM-DD] operation | agent | description
[2026-05-03] ingest | claude-code | “What is Kubernetes?” | ai-ml/nvidia-certs/ncp-aai/deployment-and-scaling
Source: raw/articles/What is Kubernetes.md | Type: article | confidence: low Tags: deployment-scaling, gpu-acceleration, multi-tenancy, inference-optimization Primary page: ai-ml/nvidia-certs/ncp-aai/deployment-and-scaling/what-is-kubernetes Entity updates: entities/nvidia.md (sources 24→33) Contradiction pre-check: no conflicts found
[2026-05-03] ingest | claude-code | “Scaling LLMs with NVIDIA Triton and NVIDIA TensorRT-LLM Using Kubernetes” | ai-ml/nvidia-certs/ncp-aai/deployment-and-scaling + ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation
Source: raw/articles/Scaling LLMs with NVIDIA Triton and NVIDIA TensorRT-LLM Using Kubernetes.md | Type: article | confidence: low Tags: deployment-scaling, inference-optimization, llm, gpu-acceleration, observability, multi-tenancy, kv-cache Primary page: ai-ml/nvidia-certs/ncp-aai/deployment-and-scaling/scaling-llms-with-nvidia-triton-and-tensorrt-llm-using-kubernetes Cross-section: ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation/scaling-llms-with-nvidia-triton-and-tensorrt-llm-using-kubernetes Entity stubs created: entities/maggie-zhang.md Contradiction pre-check: no conflicts found
[2026-05-03] ingest | claude-code | “Performance Analysis — TensorRT LLM” | ai-ml/nvidia-certs/ncp-aai/deployment-and-scaling + ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation
Source: raw/articles/Performance Analysis — TensorRT LLM.md | Type: article | confidence: low Tags: inference-optimization, deployment-scaling, observability, cuda, gpu-acceleration, llm, mixture-of-experts Primary page: ai-ml/nvidia-certs/ncp-aai/deployment-and-scaling/performance-analysis-tensorrt-llm Cross-section: ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation/performance-analysis-tensorrt-llm Contradiction pre-check: no conflicts found
[2026-05-03] ingest | claude-code | “NVIDIA Nsight Systems” | ai-ml/nvidia-certs/ncp-aai/deployment-and-scaling + ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation
Source: raw/articles/NVIDIA Nsight Systems.md | Type: article | confidence: low | Note: 485KB file due to embedded base64 PNG (line 73); actual text ~4K tokens — standard ingest applied Tags: observability, cuda, gpu-acceleration, deployment-scaling, hpc, inference-optimization Primary page: ai-ml/nvidia-certs/ncp-aai/deployment-and-scaling/nvidia-nsight-systems Cross-section: ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation/nvidia-nsight-systems Contradiction pre-check: no conflicts found
[2026-05-03] ingest | claude-code | “Measure and Improve AI Workload Performance with NVIDIA DGX Cloud Benchmarking” | ai-ml/nvidia-certs/ncp-aai/deployment-and-scaling + ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation
Source: raw/articles/Measure and Improve AI Workload Performance with NVIDIA DGX Cloud Benchmarking.md | Type: article | confidence: low Tags: deployment-scaling, inference-optimization, nvidia-nemo, mixed-precision, gpu-acceleration, llm, distributed-training, quantization Primary page: ai-ml/nvidia-certs/ncp-aai/deployment-and-scaling/measure-and-improve-ai-workload-performance-with-nvidia-dgx-cloud-benchmarking Cross-section: ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation/measure-and-improve-ai-workload-performance-with-nvidia-dgx-cloud-benchmarking Entity stubs created: entities/emily-potyraj.md Contradiction pre-check: no conflicts found
[2026-05-03] crosslink | claude-code | “deployment-and-scaling batch” | 3 links added across 3 pages
[2026-05-01] ingest | claude-code | “How to Make Your LLM More Accurate with RAG & Fine-Tuning” | ai-ml/nvidia-certs/ncp-aai/knowledge-integration-and-data-handling
Source: raw/articles/How to Make Your LLM More Accurate with RAG & Fine-Tuning.md | Type: article | Re-ingest (section correction: was ai-ml → ai-ml/nvidia-certs/ncp-aai/knowledge-integration-and-data-handling) Tags: rag, fine-tuning, llm, vector-search, embedding, langchain, lora, hallucination, peft Old page deleted: ai-ml/How to Make Your LLM More Accurate with RAG & Fine-Tuning.md New page: ai-ml/nvidia-certs/ncp-aai/knowledge-integration-and-data-handling/how-to-make-your-llm-more-accurate-with-rag-and-fine-tuning.md Entity updates: entities/sarah-schurch.md (path corrected) Cross-link updates: ai-ml/DeepSeek-R1…, ncp-aai/agent-architecture-and-design/building-autonomous-ai-nvidia-agentic-nemo.md Contradiction pre-check: no conflicts found
[2026-05-01] ingest | claude-code | “Building Autonomous AI with NVIDIA Agentic NeMo” | ai-ml/nvidia-certs/ncp-aai/agent-architecture-and-design
Source: raw/articles/Building Autonomous AI with NVIDIA Agentic NeMo.md | Type: article | Re-ingest (section correction: was ai-ml → ai-ml/nvidia-certs/ncp-aai/agent-architecture-and-design) Tags: agentic-ai, llm, rag, guardrails, agent-architecture, tool-calling, inference-optimization, llm-orchestration, state-management, lora, perceive-reason-act, nvidia-nemo, memory-augmentation, deployment-scaling Old page deleted: ai-ml/Building Autonomous AI with NVIDIA Agentic NeMo.md New page: ai-ml/nvidia-certs/ncp-aai/agent-architecture-and-design/building-autonomous-ai-nvidia-agentic-nemo.md Entity updates: entities/zia-babar.md, entities/nvidia.md (path corrected in appears_in and body links) Potential conflicts: 5-phase agentic loop (this article) vs. 4-phase PRA-L cycle (NCP-AAI materials) — compatible decompositions, noted in Internal Tensions section
[2026-04-30] tag-apply | claude-code | 2 tags added to config/tags.yml: multi-tenancy, checkpointer
[2026-04-30] tag-discovery | claude-code | ai-ml/nvidia-certs (deep_nesting scan — all depths) | 2 new candidates → config/proposed_tags.yml
Pages scanned: 84 | Existing vocabulary: 166 tags | Already decided: 0 (all prior entries approved) Deep-nesting verification: section assigned to ai-ml (tags_key), not deep path — PASS
[2026-04-30] ingest | claude-code | “Building Autonomous AI with NVIDIA Agentic NeMo” | ai-ml
Source: raw/articles/Building Autonomous AI with NVIDIA Agentic NeMo.md | Type: article | confidence: low Tags: agentic-ai, llm, rag, guardrails, agent-architecture, tool-calling, inference-optimization, llm-orchestration, state-management, lora, perceive-reason-act Entity stubs created: entities/zia-babar.md | Entity updates: entities/nvidia.md (sources 13→14) Potential conflicts: 5-phase agentic loop (this article) vs. 4-phase PRA-L cycle (NCP-AAI materials) — compatible decompositions, noted in Internal Tensions section
[2026-04-30] ingest | claude-code | “How to Make Your LLM More Accurate with RAG & Fine-Tuning” | ai-ml
Source: raw/articles/How to Make Your LLM More Accurate with RAG & Fine-Tuning.md | Type: article | confidence: low Tags: rag, fine-tuning, llm, vector-search, embedding, langchain, lora, hallucination Entity stubs created: entities/sarah-schurch.md Contradiction pre-check: no conflicts found
[2026-04-30] ingest | claude-code | “NCP-AAI Part 4 — Building Retriever Nodes: Hands-On Assessment Study Guide” | ai-ml/nvidia-certs
Source: raw/notes/NCP-AAI_Part4_Building_Retriever_Nodes_Study_Guide.docx | Type: docx | confidence: medium | Standard ingest Tags: agentic-ai, rag, structured-output, pydantic, prompt-engineering, llm, langchain, python, tool-calling, vector-search Entity updates: entities/nvidia.md (sources 12→13) Contradiction pre-check: no conflicts found
[2026-04-29] ingest | claude-code | “Training Compute-Optimal Large Language Models (Chinchilla)” | ai-ml
Source: raw/papers/Training Compute-Optimal Large Language Models.pdf | Type: paper | confidence: medium | Staged ingest Grouping: LLM-assisted | 2 groups from 6 chapter pages Group 1 — Core paper: Ch. 1–5 (Introduction, Related Work, Methods, Chinchilla, Discussion) Group 2 — Appendices: Ch. 6 (A–J combined) Entity stubs created: deepmind, jordan-hoffmann, sebastian-borgeaud, arthur-mensch, laurent-sifre Quality flag: tag “scaling-laws” missing from config/tags.yml — run tag-discovery
[2026-04-29] lint | claude-code | wireless/isac | 1 issue found → scratch/lint-2026-04-29.md
Checks: 1–11 local | Pages scanned: 2 | [check 10] query page how-do-the-ml-models-used-in-wireless-sensing-cnns-in-80211b.md — tools/NCP-AAI in sources_consulted last_ingested 2026-04-28 > query_date 2026-04-13
[2026-04-27] tag-apply | claude-code | 20 tags added to config/tags.yml: outage-probability, water-filling, spectral-efficiency, rayleigh-fading, power-control, sinr, channel-state-information, intersymbol-interference, spread-spectrum, cdma, path-loss, shadowing, delay-spread, coherence-time, diversity-order, maximal-ratio-combining, successive-interference-cancellation, multiuser-diversity, dirty-paper-coding, frequency-selective-fading
[2026-04-27] tag-discovery | claude-code | wireless/ (all subsections) | 20 new candidates → config/proposed_tags.yml
Pages scanned: 26/26 | Existing vocabulary: 38 wireless tags | Already decided: 0
[2026-04-26] lint | claude-code | ai-ml/ (full section, 37 pages) | 2 issues found → scratch/lint-2026-04-26.md
Checks: 1–11 local + 12–13 cloud | Tokens: ~43,000 in / ~1,800 out | Est. cost: ~$0.16
[2026-04-26] lint | claude-code | ai-ml/nvidia-certs/ (full section, 18 pages) | 6 issues found → scratch/lint-2026-04-26.md
[2026-04-24] ingest | claude-code | “NVIDIA DLI: Building Agentic AI Applications with LLMs” | ai-ml
Source: raw/notes/NCP-AAI_Part2_Exam_Prep_Full.docx | Type: note | confidence: medium
[2026-04-13] ingest | claude-code | “NCP-AAI CERTIFICATION Part 0 - Foundations and Responsible AI” | ai-ml
Source: raw/notes/NCP-AAI_Part0_Exam_Prep_FULL.docx | Type: note
[2026-04-13] ingest | claude-code | “NCP-AAI Exam Preparation” | ai-ml + tools
Source: raw/notes/NCP-AAI_Part_1_Exam_Prep_FULL.docx | Type: note | Cross-section: 2 pages written
[2026-04-13] ingest | claude-code | “Building Agentic AI Applications with LLMs Part 3: Graph-Based Orchestration” | ai-ml
Source: raw/notes/NCP-AAI_Part3_GraphBased_Orchestration_Study_Guide.docx | Type: note
[2026-04-13] ingest | claude-code | “Attention Is All You Need” | ai-ml
Source: raw/papers/NIPS-2017-attention-is-all-you-need-Paper.pdf | Type: paper
[2026-04-13] ingest | claude-code | “5G PRS-Based Sensing: A Sensing Reference Signal Approach for ISAC” | wireless/isac + wireless/5g-nr
Source: raw/papers/5G PRS based sening a sensing reference signal approach for ISAC.pdf | Type: paper | confidence: medium | Cross-section: 2 pages written
[2026-04-13] ingest | claude-code | “802.11bf Multiband Passive Sensing: Reusing Wi-Fi Signaling for Sensing” | wireless/isac + ai-ml
Source: raw/papers/802.11bf multiband passive sening reusing wifi singling for sensing.pdf | Type: paper | confidence: medium | Cross-section: 2 pages written
[2026-04-13] query | claude-code | “How do the ML models used in wireless sensing (CNNs in 802.11bf) compare to tran” | ., ai-ml, finance, tools, wireless, wireless/5g-nr, wireless/isac
Tokens: 12,369 in / 1,191 out | Est. cost: ~$0.0550 | Saved as: wiki/queries/how-do-the-ml-models-used-in-wireless-sensing-cnns-in-80211b.md
[2026-04-13] lint | claude-code | all sections | 0 issues found → wiki/scratch/lint-2026-04-13.md
[2026-04-13] ingest | claude-code | “Bitcoin: A Peer-to-Peer Electronic Cash System” | finance
Source: raw/papers/bitcoin.pdf | Type: paper
[2026-04-13] ingest | claude-code | “What Is Kalshi? A Beginner’s Guide” | finance
Source: raw/articles/What Is Kalshi_ A Beginner’s Guide.md | Type: article | confidence: low
[2026-04-13] lint | claude-code | finance/ | 0 issues found → wiki/scratch/lint-2026-04-13.md
[2026-04-13] lint | claude-code | wireless, ai-ml, tools/ | 0 issues found → wiki/scratch/lint-2026-04-13.md
[2026-04-13] lint | claude-code | finance/ | 1 issue found → wiki/scratch/lint-2026-04-13.md
[2026-04-13] lint | claude-code | finance/ | 2 issues found → wiki/scratch/lint-2026-04-13.md
[2026-04-13] lint | claude-code | finance/ | 2 issues found → wiki/scratch/lint-2026-04-13.md
[2026-04-13] lint | claude-code | finance/ | 2 issues found → wiki/scratch/lint-2026-04-13.md
[2026-04-13] lint | claude-code | finance/ | 1 issue found → wiki/scratch/lint-2026-04-13.md
[2026-04-13] lint | claude-code | ai-ml/ | 0 issues found → wiki/scratch/lint-2026-04-13.md
[2026-04-13] lint | claude-code | ai-ml/ | 2 issues found → wiki/scratch/lint-2026-04-13.md
[2026-04-13] lint | claude-code | all sections | 4 issues found → wiki/scratch/lint-2026-04-13.md
[2026-04-13] lint | claude-code | all sections | 4 issues found → wiki/scratch/lint-2026-04-13.md
[2026-04-13] lint | claude-code | all sections | 0 issues found → wiki/scratch/lint-2026-04-13.md
[2026-04-14] ingest | claude-code | “Wireless Communications (2nd ed. Draft) — Andrea Goldsmith” | wireless/signal-processing + wireless/5g-nr
Source: raw/papers/WirelessComm_Chp1-16_March32020.pdf | Type: pdf | Staged ingest | Cross-section: 2 pages written Grouping: hybrid | 4 groups from 16 chapters Group 1 — Foundations & Channel Models: ch. 1–4 | 1,663 tokens Group 2 — Link Reliability & Transmission: ch. 5–9 | 2,213 tokens Group 3 — Physical Techniques & Access: ch. 10–13 | 1,580 tokens Group 4 — Systems & Networks: ch. 14–16 | 1,475 tokens Tokens: 2,790 in / 3,000 out | Est. cost: ~$0.0534
[2026-04-15] ingest | claude-code | “Building Agentic AI Applications with LLMs” | ai-ml
Source: raw/slides/Building_Agentic_AI_Applications_with_LLMs.pptx | Type: slides | Staged ingest Grouping: token_budget | 1 groups from 1 chapters Group 1 — Building Agentic AI Applications: ch. 1–1 | 432 tokens Tokens: 414 in / 3,000 out | Est. cost: ~$0.0462
[2026-04-15] ingest | claude-code | “Building Agentic AI Applications with LLMs” | ai-ml
Source: raw/slides/Building_Agentic_AI_Applications_with_LLMs.pptx | Type: slides | Staged ingest Grouping: hybrid | 3 groups from 14 chapters Group 1 — Foundations and Basic Agents: ch. 1–6 | 2,433 tokens Group 2 — Control, Structure, and Tooling: ch. 7–12 | 1,358 tokens Group 3 — Advanced Notebooks and Applications: ch. 13–14 | 1,022 tokens Tokens: 1,742 in / 3,000 out | Est. cost: ~$0.0502
[2026-04-24] ingest | claude-code | “DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning” | ai-ml
Source: raw/papers/DeepSeek-R1 Incentivizing Reasoning Capability in LLMs via Reinforcement Learing.pdf | Type: paper
[2026-04-24] ingest | claude-code | “Generative AI LLM Exam Study Guide” | ai-ml
Source: raw/slides/Generative AI LLM Exam Study Guide.pptx | Type: slides | Staged ingest Grouping: hybrid | 3 groups from 9 chapters Group 1 — Core ML, Data Science & Experimentation: ch. 1–3 | 2,428 tokens Group 2 — LLM Training, Customization & Inference: ch. 4–6 | 2,780 tokens Group 3 — Deployment, RAG & Trustworthy AI: ch. 7–9 | 3,058 tokens Tokens: 2,306 in / 3,000 out | Est. cost: ~$0.0519
[2026-04-24] lint | claude-code | all sections | 2 issues found → wiki/scratch/lint-2026-04-24.md
[2026-04-25] ingest | claude-code | “Wireless Communications (2nd ed. Draft) — Andrea Goldsmith” | wireless/signal-processing
Source: raw/papers/WirelessComm_Chp1-16_March32020.pdf | Type: pdf | Staged ingest Grouping: hybrid | 6 groups from 16 chapters Group 1 — Wireless Channel & Overview: ch. 1–3 | 8,118 tokens Group 2 — Channel Capacity & Digital Modulation: ch. 4–6 | 8,480 tokens Group 3 — Transmission Reliability & Adaptation: ch. 7–9 | 7,441 tokens Group 4 — MIMO & Advanced Modulation: ch. 10–12 | 8,413 tokens Group 5 — Multiuser & Spread Spectrum: ch. 13–14 | 5,732 tokens Group 6 — Wireless Network Architectures: ch. 15–16 | 4,904 tokens Pages written: 1 synthesis + 16 chapter pages Tokens: 4,798 in / 3,000 out | Est. cost: ~$0.0594
[2026-04-25] manual | claude-code | “Create ai-ml/nvidia-certs subsection” | ai-ml
Moved 5 pages from ai-ml/ to ai-ml/nvidia-certs/: NCP-AAI_Part0, Part1, Part2, Part3, Generative AI LLM Exam Study Guide Updated: section/primary_section frontmatter on moved pages; tools/NCP-AAI_Part_1 cross-section page; entities/nvidia, entities/russell-norvig; cross-links in 4 other pages Created: wiki/ai-ml/nvidia-certs/index.md; restructured wiki/ai-ml/index.md with subsection entry CLAUDE.md v1.11: ai-ml Section 10 updated with nvidia-certs description
[2026-04-25] ingest | claude-code | “Generative AI LLM Exam Study Guide” | ai-ml/nvidia-certs
Source: raw/slides/Generative AI LLM Exam Study Guide.pptx | Type: slides | Staged ingest Grouping: hybrid | 3 groups from 9 chapters Group 1 — Core ML & Experimentation: ch. 1–3 | 6,814 tokens Group 2 — LLM Fundamentals & Optimization: ch. 4–6 | 7,829 tokens Group 3 — Advanced Applications & Trustworthy AI: ch. 7–9 | 6,948 tokens Pages written: 1 synthesis + 9 chapter pages Tokens: 2,457 in / 3,000 out | Est. cost: ~$0.0524
[2026-04-25] ingest | claude-code | “Building Agentic AI Applications with LLMs” | ai-ml
Source: raw/slides/Building_Agentic_AI_Applications_with_LLMs.pptx | Type: slides | Staged ingest Grouping: hybrid | 4 groups from 14 chapters Group 1 — Foundations & First Agent Build: ch. 1–4 | 8,395 tokens Group 2 — Agent Frameworks & Output Structuring: ch. 5–8 | 8,693 tokens Group 3 — Advanced Tooling & Data Systems: ch. 9–12 | 8,579 tokens Group 4 — Notebook Implementations: ch. 13–14 | 5,152 tokens Pages written: 1 synthesis + 14 chapter pages Tokens: 2,778 in / 3,000 out | Est. cost: ~$0.0533
[2026-04-26] ingest | claude-code | “Generative AI with Diffusion Models” | ai-ml/nvidia-certs
Source: raw/slides/NCA-GENM Generative AI with Diffusion Models.pptx | Type: slides
[2026-04-26] ingest | claude-code | “Multimodal Data” | ai-ml/nvidia-certs
Source: raw/slides/NCA-GENM Multimodal Data.pptx | Type: slides
[2026-04-26] ingest | claude-code | “Software Development” | ai-ml/nvidia-certs
Source: raw/slides/NCA-GENM Softerware development.pptx | Type: slides
[2026-04-26] lint | claude-code | ai-ml/nvidia-certs (3 NCA-GENM pages) | 2 issues found → scratch/lint-2026-04-26.md
[2026-04-27] ingest | claude-code | “Core Machine Learning and AI Knowledge” | ai-ml/nvidia-certs
Source: raw/slides/NCA-GENM Core Machine Learning and AI Knowledge.pptx | Type: slides
[2026-04-27] ingest | claude-code | “Experimentation” | ai-ml/nvidia-certs
Source: raw/slides/NCA-GENM Experimentation.pptx | Type: slides | Staged ingest Grouping: hybrid | 3 groups from 9 chapters Group 1 — AI Fundamentals & Data Visualization: ch. 1–3 | 7,678 tokens Group 2 — Data Preprocessing & Cleaning: ch. 4–6 | 5,745 tokens Group 3 — Advanced AI Modeling & Ethics: ch. 7–9 | 7,375 tokens Pages written: 1 synthesis + 9 chapter pages Tokens: 2,201 in / 3,000 out | Est. cost: ~$0.0516
[2026-04-27] ingest | claude-code | “Performance Optimization” | ai-ml/nvidia-certs
Source: raw/slides/NCA-GENM Performance Optimization.pptx | Type: slides
[2026-04-27] tag-discovery | claude-code | ai-ml/ (all subsections) | 31 new candidates → config/proposed_tags.yml
Pages scanned: 42/49 | Existing vocabulary: 34 ai-ml tags | Already decided: 0
[2026-04-27] tag-apply | claude-code | 31 tags added to config/tags.yml: diffusion-model, multimodal-learning, data-flywheel, generative-adversarial-network, self-attention, distributed-training, vision-transformer, contrastive-learning, convolutional-neural-network, context-window, test-time-compute, lora, kv-cache, hallucination, catastrophic-forgetting, knowledge-distillation, data-augmentation, speech-recognition, zero-shot-learning, embedding, positional-encoding, regularization, encoder-decoder, llmops, vector-search, autoencoder, u-net, mixed-precision, data-preprocessing, dimensionality-reduction, flash-attention
[2026-04-28] ingest | claude-code | “Document Overview” | ai-ml/nvidia-certs
Source: raw/notes/NCP-AAI_Part0_Exam_Prep_FULL.md | Type: docx | Staged ingest Grouping: hybrid | 3 groups from 10 chapters Group 1 — Introduction to Agentic AI & Architecture: ch. 1–4 | 9,239 tokens Group 2 — Agent Types, Ethics & Resources: ch. 5–7 | 6,753 tokens Group 3 — Practice Assessment & Quick Reference: ch. 8–10 | 5,764 tokens Pages written: 1 synthesis + 10 chapter pages Tokens: 2,220 in / 3,000 out | Est. cost: ~$0.0517
[2026-04-28] ingest | claude-code | “Document Overview” | ai-ml/nvidia-certs + tools
Source: raw/notes/NCP-AAI_Part_1_Exam_Prep_FULL.md | Type: docx | Staged ingest | Cross-section: 2 pages written Grouping: hybrid | 4 groups from 10 chapters Group 1 — Overview & Deep Learning Foundations: ch. 1–2 | 4,527 tokens Group 2 — Architecture & Semantic Reasoning: ch. 3–4 | 5,570 tokens Group 3 — Agents, Frameworks & Context Limits: ch. 5–7 | 7,476 tokens Group 4 — Assessment & Quick Reference: ch. 8–10 | 8,476 tokens Pages written: 1 synthesis + 10 chapter pages Tokens: 3,021 in / 3,000 out | Est. cost: ~$0.0541
[2026-04-28] ingest | claude-code | “NVIDIA DLI: Building Agentic AI Applications with LLMs” | ai-ml/nvidia-certs
Source: raw/notes/NCP-AAI_Part2_Exam_Prep_Full.md | Type: docx | Staged ingest Grouping: hybrid | 3 groups from 8 chapters Group 1 — Course Introduction and Foundational Concepts: ch. 1–3 | 8,197 tokens Group 2 — Assessment and Practice Materials: ch. 4–6 | 7,393 tokens Group 3 — Quick Reference and Study Strategy: ch. 7–8 | 4,830 tokens Pages written: 1 synthesis + 8 chapter pages Tokens: 2,356 in / 3,000 out | Est. cost: ~$0.0521
[2026-04-28] ingest | claude-code | “Table of Contents” | ai-ml/nvidia-certs
Source: raw/notes/NCP-AAI_Part3_GraphBased_Orchestration_Study_Guide.md | Type: docx | Staged ingest Grouping: hybrid | 4 groups from 13 chapters Group 1 — Introduction & Conceptual Foundation: ch. 1–4 | 9,576 tokens Group 2 — Architecture, Patterns, and Framework Design: ch. 5–8 | 9,929 tokens Group 3 — Troubleshooting and Assessment: ch. 9–10 | 4,469 tokens Group 4 — Reference Materials and Appendices: ch. 11–13 | 7,905 tokens Pages written: 1 synthesis + 13 chapter pages Tokens: 3,173 in / 3,000 out | Est. cost: ~$0.0545
[2026-04-29] ingest | claude-code | “DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning” | ai-ml
Source: raw/papers/DeepSeek-R1 Incentivizing Reasoning Capability in LLMs via Reinforcement Learing.pdf | Type: pdf | Staged ingest Grouping: hybrid | 5 groups from 17 chapters Group 1 — Introduction and Core Model Overview: ch. 1–4 | 9,935 tokens Group 2 — Ethics, Conclusion, and Background: ch. 5–8 | 9,885 tokens Group 3 — Training Details and Self-Evolution: ch. 9–12 | 9,704 tokens Group 4 — Analysis and Related Work: ch. 13–15 | 6,538 tokens Group 5 — Open Source, Code, and Evaluation Settings: ch. 16–17 | 4,110 tokens Pages written: 1 synthesis + 17 chapter pages Tokens: 3,862 in / 3,000 out | Est. cost: ~$0.0566
[2026-04-29] tag-discovery | claude-code | ai-ml (DeepSeek-R1 re-ingest) | 8 new → config/proposed_tags.yml
[2026-04-29] tag-apply | claude-code | 8 tags added to config/tags.yml: rejection-sampling, reward-hacking, proximal-policy-optimization, jailbreaking, mixture-of-experts, monte-carlo-tree-search, llm-as-a-judge, cold-start
[2026-04-30] lint | claude-code | ai-ml/nvidia-certs/ | 7 issues found → scratch/lint-2026-04-30.md
[2026-04-30] manual | claude-code | “Remove stale query page” | queries/how-do-the-ml-models-used-in-wireless-sensing-cnns-in-80211b.md — 4 nvidia-certs sources re-ingested after query_date 2026-04-13
[2026-05-01] ingest | claude-code | “What are Multi-Agent Systems?” | ai-ml/nvidia-certs/ncp-aai/agent-architecture-and-design
Source: raw/articles/What are Multi-Agent Systems.md | Type: article | Standard ingest
[2026-05-01] ingest | claude-code | “Three Building Blocks for Creating AI Virtual Assistants for Customer Service with an NVIDIA AI Blueprint” | ai-ml/nvidia-certs/ncp-aai/agent-architecture-and-design
Source: raw/articles/Three Building Blocks for Creating AI Virtual Assistants for Customer Service with an NVIDIA AI Blueprint.md | Type: article | Standard ingest
[2026-05-01] ingest | claude-code | “Catch Me If You Can: A Multi-Agent Framework for Financial Fraud Detection” | ai-ml/nvidia-certs/ncp-aai/agent-architecture-and-design
Source: raw/articles/Catch Me If You Can_ A Multi-Agent Framework for Financial Fraud Detection.md | Type: article | Standard ingest
[2026-05-01] ingest | claude-code | “What Is Agent Memory? A Guide to Enhancing AI Learning and Recall” | ai-ml/nvidia-certs/ncp-aai/cognition-planning-and-memory
Source: raw/articles/What Is Agent Memory_ A Guide to Enhancing AI Learning and Recall.md | Type: article | Standard ingest
[2026-05-02] ingest | claude-code | “Optimization — NVIDIA Triton Inference Server” | ai-ml/nvidia-certs/ncp-aai/agent-development + deployment-and-scaling + nvidia-platform-implementation
Source: raw/articles/Optimization — NVIDIA Triton Inference Server.md | Type: article | Standard ingest | Cross-section: 2 pages written
[2026-05-02] ingest | claude-code | “An Introduction to Large Language Models: Prompt Engineering and P-Tuning” | ai-ml/nvidia-certs/ncp-aai/agent-development + cognition-planning-and-memory
Source: raw/articles/An Introduction to Large Language Models_ Prompt Engineering and P-Tuning.md | Type: article | Standard ingest | Cross-section: 1 page written
[2026-05-02] ingest | claude-code | “Circuit Breaker Pattern” | ai-ml/nvidia-certs/ncp-aai/agent-development
Source: raw/articles/Circuit Breaker Pattern - Azure Architecture Center.md | Type: article | Standard ingest
[2026-05-02] ingest | claude-code | “Design Considerations of Advanced Agentic AI for Real-World Applications” | ai-ml/nvidia-certs/ncp-aai/agent-development + agent-architecture-and-design
Source: raw/articles/Design Considerations of Advanced Agentic AI for Real world Applications.md | Type: article | Standard ingest | Cross-section: 1 page written
[2026-05-02] ingest | claude-code | “Transient Fault Handling — Best Practices” | ai-ml/nvidia-certs/ncp-aai/agent-development
Source: raw/articles/Transient Fault Handling - Azure Architecture Center.md | Type: article | Standard ingest
[2026-05-02] ingest | claude-code | “Retry Pattern” | ai-ml/nvidia-certs/ncp-aai/agent-development
Source: raw/articles/Retry pattern - Azure Architecture Center.md | Type: article | Standard ingest
[2026-05-02] lint | claude-code | ai-ml/nvidia-certs/ncp-aai/ | 6 issues found → scratch/lint-2026-05-02.md
[2026-05-02] manual | claude-code | fix lint issues 1-6 in ai-ml/nvidia-certs/ncp-aai/
Issue 1: created entities/crewai.md stub; fixed bad link in ncp-aai-agent-arch-test-note (→ entities/crewai) Issues 2-3: added 2 Triton cross-section pages to entities/nvidia.md appears_in (sources 18→20) Issue 4: added RAG article to entities/langchain.md appears_in (sources 2→3) Issue 5: created entities/indrajit-kar.md stub (sources: 2); linked author in Design-Considerations primary page Issue 6: created entities/tanay-varshney.md stub (sources: 2); linked author in An-Introduction-to-LLMs primary page
[2026-05-03] ingest | claude-code | “What are AI Agents?” | ai-ml/nvidia-certs/ncp-aai/agent-architecture-and-design
Source: raw/articles/What are AI Agents.md | Type: article | Standard ingest
[2026-05-03] ingest | claude-code | “Data Flywheel: What It Is and How It Works” | ai-ml/nvidia-certs/ncp-aai/evaluation-and-tuning
Source: raw/articles/Data flywheel_ What it is and how it works.md | Type: article | Standard ingest
[2026-05-03] ingest | claude-code | “AI Agents in Production: Observability & Evaluation” | ai-ml/nvidia-certs/ncp-aai/evaluation-and-tuning
Source: raw/articles/AI Agents in Production_ Observability & Evaluation.md | Type: article | Standard ingest
[2026-05-03] ingest | claude-code | “NVIDIA NeMo Agent Toolkit: Agent Evaluation” | ai-ml/nvidia-certs/ncp-aai/evaluation-and-tuning + nvidia-platform-implementation
Source: raw/articles/NVIDIA_NeMo-Agent-Toolkit_ The NVIDIA NeMo Agent toolkit is an open-source library for efficiently connecting and optimizing teams of AI agents..md | Type: article | Standard ingest | Cross-section: 1 page written
[2026-05-03] ingest | claude-code | “Navigating the Challenges: 5 Common Pitfalls in Agentic AI Adoption” | ai-ml/nvidia-certs/ncp-aai/safety-ethics-and-compliance
Source: raw/articles/Navigating the Challenges_ 5 Common Pitfalls in Agentic AI.md | Type: article | Standard ingest
[2026-05-03] ingest | claude-code | “Successful Agentic AI: Model Logic, Data Considerations and Manpower” | ai-ml/nvidia-certs/ncp-aai/evaluation-and-tuning
Source: raw/articles/Successful Agentic AI_ Model Logic, Data Considerations and Manpower.md | Type: article | Standard ingest
[2026-05-03] ingest | claude-code | “Are Large Language Models In-Context Graph Learners?” | ai-ml/nvidia-certs/ncp-aai/cognition-planning-and-memory
Source: raw/papers/Are Large Language Models In-Context Graph Learners.pdf | Type: paper
[2026-05-03] ingest | claude-code | “From Human Memory to AI Memory: A Survey on Memory Mechanisms in the Era of LLMs” | ai-ml/nvidia-certs/ncp-aai/cognition-planning-and-memory
Source: raw/papers/From Human Memory to AI Memory.pdf | Type: paper
[2026-05-03] ingest | claude-code | “Understanding the planning of LLM agents: A survey” | ai-ml/nvidia-certs/ncp-aai/cognition-planning-and-memory
Source: raw/papers/Understanding the planning of LLM agents A survey.pdf | Type: pdf | Staged ingest Grouping: hybrid | 4 groups from 9 chapters Group 1 — Introduction and Taxonomy: ch. 1–2 | 5,675 tokens Group 2 — Task Decomposition and Multi-Plan Selection: ch. 3–4 | 5,808 tokens Group 3 — External, Reflective, and Memory-Augmented Planning: ch. 5–7 | 7,071 tokens Group 4 — Evaluation and Conclusions: ch. 8–9 | 4,555 tokens Pages written: 1 synthesis + 9 chapter pages Tokens: 3,279 in / 3,000 out | Est. cost: ~$0.0548
[2026-05-03] ingest | claude-code | “Jamba 1.5 LLMs Leverage Hybrid Architecture to Deliver Superior Reasoning and Long Context Handling” | ai-ml/nvidia-certs/ncp-aai/cognition-planning-and-memory
Source: raw/articles/Jamba 1.5 LLMs Leverage Hybrid Architecture to Deliver Superior Reasoning and Long Context Handling.md | Type: article
[2026-05-04] ingest | claude-code | “Chat With Your Enterprise Data Through Open-Source AI-Q NVIDIA Blueprint” | ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation + knowledge-integration-and-data-handling
Source: raw/articles/Chat With Your Enterprise Data Through Open-Source AI-Q NVIDIA Blueprint.md | Type: article | Standard ingest | Cross-section: 1 page written
[2026-05-04] ingest | claude-code | “Improve AI Code Generation Using NVIDIA NeMo Agent Toolkit” | ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation + agent-development
Source: raw/articles/Improve AI Code Generation Using NVIDIA NeMo Agent Toolkit.md | Type: article | Standard ingest | Cross-section: 1 page written
[2026-05-04] ingest | claude-code | “NVIDIA NeMo Agent Toolkit” | ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation
Source: raw/articles/NVIDIA NeMo Agent Toolkit.md | Type: article | Standard ingest
[2026-05-04] ingest | claude-code | “Measure and Improve AI Workload Performance with NVIDIA DGX Cloud Benchmarking” | SKIPPED — already ingested 2026-05-03
[2026-05-04] ingest | claude-code | “Welcome to NVIDIA Run:ai Documentation” | ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation + deployment-and-scaling
Source: raw/articles/Welcome to NVIDIA Run_ai Documentation.md | Type: article | Standard ingest | Cross-section: 1 page written
[2026-05-04] ingest | claude-code | “Performance Tuning Guide — Megatron-Bridge LLM Training” | ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation + deployment-and-scaling
Source: raw/articles/Performance Tuning Guide.md | Type: article | Standard ingest | Cross-section: 1 page written
[2026-05-04] ingest | claude-code | “NVIDIA NeMo Guardrails” | ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation + safety-ethics-and-compliance
Source: raw/articles/NVIDIA NeMo Guardrails.md | Type: article | Standard ingest | Cross-section: 1 page written
[2026-05-04] ingest | claude-code | “Triton Inference Server Backend” | ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation + deployment-and-scaling
Source: raw/articles/Triton Inference Server Backend — NVIDIA Triton Inference Server.md | Type: article | Standard ingest | Cross-section: 1 page written
[2026-05-04] ingest | claude-code | “Batchers — NVIDIA Triton Inference Server” | ai-ml/nvidia-certs/ncp-aai/nvidia-platform-implementation + deployment-and-scaling
Source: raw/articles/Batchers — NVIDIA Triton Inference Server.md | Type: article | Standard ingest | Cross-section: 1 page written
[2026-05-04] ingest | claude-code | “AI Agent Evaluation — Summary” | ai-ml/nvidia-certs/ncp-aai/run-monitor-and-maintain + evaluation-and-tuning
Source: raw/articles/ai_agent_evaluation_summary.md | Type: article | Standard ingest | Cross-section: 1 page written
[2026-05-04] ingest | claude-code | “Log, Trace, and Monitor Portkey Integrations” | ai-ml/nvidia-certs/ncp-aai/run-monitor-and-maintain
Source: raw/articles/Log, trace, and monitor Portkey integrations.md | Type: article | Standard ingest
[2026-05-04] ingest | claude-code | “Time-Weighted Retrieval Integration” | ai-ml/nvidia-certs/ncp-aai/run-monitor-and-maintain
Source: raw/articles/Time-weighted integration.md | Type: article | Standard ingest
[2026-05-04] ingest | claude-code | “Observability Concepts (LangSmith)” | ai-ml/nvidia-certs/ncp-aai/run-monitor-and-maintain
Source: raw/articles/Observability concepts.md | Type: article | Standard ingest
[2026-05-04] ingest | claude-code | “Structured Output (LangChain Agents)” | ai-ml/nvidia-certs/ncp-aai/run-monitor-and-maintain + agent-development
Source: raw/articles/Structured output.md | Type: article | Standard ingest | Cross-section: 1 page written
[2026-05-04] ingest | claude-code | “How to Handle Model Rate Limits” | ai-ml/nvidia-certs/ncp-aai/run-monitor-and-maintain
Source: raw/articles/How to handle model rate limits.md | Type: article | Standard ingest
[2026-05-04] ingest | claude-code | “Troubleshooting — TensorRT-LLM” | ai-ml/nvidia-certs/ncp-aai/evaluation-and-tuning + nvidia-platform-implementation
Source: raw/articles/Troubleshooting — TensorRT LLM.md | Type: article | Standard ingest | Cross-section: 1 page written
[2026-05-04] ingest | claude-code | “A Guide to Monitoring Machine Learning Models in Production” | ai-ml/nvidia-certs/ncp-aai/evaluation-and-tuning + run-monitor-and-maintain
Source: raw/articles/A Guide to Monitoring Machine Learning Models in Production.md | Type: article | Standard ingest | Cross-section: 1 page written
[2026-05-04] ingest | claude-code | “Monitoring ML Models in Production: Data Quality and Integrity” | ai-ml/nvidia-certs/ncp-aai/evaluation-and-tuning + run-monitor-and-maintain
Source: raw/articles/Monitoring Machine Learning Models in Production_ How to Track Data Quality and Integrity_.md | Type: article | Standard ingest | Cross-section: 1 page written
[2026-05-04] ingest | claude-code | “Building Safer LLM Apps with LangChain Templates and NVIDIA NeMo Guardrails” | ai-ml/nvidia-certs/ncp-aai/safety-ethics-and-compliance
Source: raw/articles/Building Safer LLM Apps with LangChain Templates and NVIDIA NeMo Guardrails.md | Type: article | confidence: low
[2026-05-04] ingest | claude-code | “Agentic or Tool use” | ai-ml/nvidia-certs/ncp-aai/safety-ethics-and-compliance
Source: raw/articles/Agentic or Tool use.md | Type: article | confidence: medium
[2026-05-04] ingest | claude-code | “Artificial Intelligence in Software” | ai-ml/nvidia-certs/ncp-aai/safety-ethics-and-compliance
Source: raw/articles/Artificial Intelligence in Software.md | Type: article | confidence: low
[2026-05-04] ingest | claude-code | “Securing Generative AI Deployments with NVIDIA NIM and NVIDIA NeMo Guardrails” | ai-ml/nvidia-certs/ncp-aai/safety-ethics-and-compliance
Source: raw/articles/Securing Generative AI Deployments with NVIDIA NIM and NVIDIA NeMo Guardrails.md | Type: article | confidence: low
[2026-05-04] ingest | claude-code | “Chain of Thought Prompting Explained (with examples)” | ai-ml/nvidia-certs/ncp-aai/human-ai-interaction-and-oversight
Source: raw/articles/Chain of Thought Prompting Explained (with examples).md | Type: article | Standard ingest
[2026-05-04] ingest | claude-code | “Understanding Why AI Guardrails Are Necessary: Ensuring Ethical and Responsible AI Use” | ai-ml/nvidia-certs/ncp-aai/human-ai-interaction-and-oversight
Source: raw/articles/Understanding Why AI Guardrails Are Necessary_ Ensuring Ethical and Responsible AI Use.md | Type: article | Standard ingest
[2026-05-04] ingest | claude-code | “Human in the Loop AI: Keeping AI Aligned with Human Values” | ai-ml/nvidia-certs/ncp-aai/human-ai-interaction-and-oversight
Source: raw/articles/Human in the Loop AI_ Keeping AI Aligned with Human Values.md | Type: article | Standard ingest
[2026-05-27] manual | claude-code | “Add ai-accelerator-architectures subsection with low-latency-llm-inference topic” | ai-ml/ai-accelerator-architectures
Created: wiki/ai-ml/ai-accelerator-architectures/index.md, wiki/ai-ml/ai-accelerator-architectures/low-latency-llm-inference/index.md Updated: wiki/ai-ml/index.md, config/subjects.yml, config/tags.yml (12 new tags added)
[2026-05-27] ingest | claude-code | “Mastering Tensor Dimensions in Transformers” | ai-ml/ai-accelerator-architectures/low-latency-llm-inference
Source: raw/articles/Mastering Tensor Dimensions in Transformers.md | Type: article | Standard ingest
[2026-05-27] ingest | claude-code | “KV Caching Explained: Optimizing Transformer Inference Efficiency” | ai-ml/ai-accelerator-architectures/low-latency-llm-inference
Source: raw/articles/KV Caching Explained_ Optimizing Transformer Inference Efficiency.md | Type: article | Standard ingest
[2026-05-27] ingest | claude-code | “Optimizing Inference for Long Context and Large Batch Sizes with NVFP4 KV Cache” | ai-ml/ai-accelerator-architectures/low-latency-llm-inference
Source: raw/articles/Optimizing Inference for Long Context and Large Batch Sizes with NVFP4 KV Cache.md | Type: article | Standard ingest
[2026-05-27] ingest | claude-code | “Harness, Scaffold, and the AI Agent Terms Worth Getting Right” | ai-ml/nvidia-certs/ncp-aai/agent-architecture-and-design
Source: raw/articles/Harness, Scaffold, and the AI Agent Terms Worth Getting Right.md | Type: article | Standard ingest