Data & AI Intelligence¶
Estimated time to read: 5 minutes
The Data & AI Intelligence library covers the intersection of modern data analytics, machine learning engineering, and generative AI. It is split into two primary areas: securing autonomous AI agentic workflows (including Model Context Protocol and adversarial resilience) and structuring enterprise data pipelines using BigQuery and DBT.
Intelligence Topics¶
Select a category below to explore specific data and AI guides:
Agentic Security & Adversarial AI¶
Secure autonomous agents, prompt frameworks, and multi-agent platforms from exploits:
-
LLM Agent Vulnerability Taxonomy
A systematic breakdown of vulnerabilities in LLM agent setups, including tool execution and memory injection.
-
Hacking the Agentic Enterprise
Deconstruct attack vectors used to exploit autonomous corporate systems and LLM integrations.
-
Securing the Agentic Enterprise
Defensive strategies, guardrails, and architectures to protect agent deployments.
-
Securing AI Agents (Access Control)
Why traditional access control policies are dead and how cryptographically verified authorization must replace them.
-
Securing Agentic AI & Model Context Protocol
How to configure, monitor, and restrict integrations using the Model Context Protocol (MCP) to prevent data leakage.
-
Are Guard Models Enough?
An analysis of Llama Guard and input/output checkers, evaluating their limitations in defending agent tools.
-
Memory & Context Poisoning Attacks
Deconstruct persistence exploits where attackers insert malicious instructions into database memory or vector stores.
-
Multi-Agent Token Security
Detail token abuse and privilege escalation risks in distributed multi-agent frameworks.
-
LLM Supply Chain Risks
Understand the security implications of third-party models, compromised training data, and pipeline components.
-
OWASP Top 10 for LLMs
A reference walkthrough of the OWASP Top 10 vulnerabilities affecting Large Language Model integrations.
BigQuery & Data Pipelines¶
Leverage enterprise data warehouses and structure pipelines using DBT:
-
BigQuery Cheat Sheet
A quick command-line and SQL syntax reference for running queries, loading data, and structuring tables in BigQuery.
-
BigQuery for Data Analysts
An analyst's guide to writing SQL queries, joining tables, and generating business analytics in BigQuery.
-
BigQuery Getting Started
Learn database setup, access configurations, dataset schema definitions, and table partition models.
-
BigQuery Best Practices
Optimize query cost, run partitions/clustering, and structure storage strategies for large tables.
-
DBT Cheat Sheet
Define schemas, run materializations, test assertions, and build model lineages using DBT.
Machine Learning & AI Architectures¶
Design, build, and deploy production ML models:
-
:material-server-cog: MLOps Fundamentals
Build model tracking, lineage versioning, data registries, and deployment loops.
-
:material-cpu: Architectural Evolution of LLMs
A technical review of language model architectures, from Recurrent Nets and transformers to modern reasoning loops.
-
BigQuery ML for Starters
Build, train, and run prediction models (linear, log, k-means) directly in BigQuery using standard SQL.
-
BigQuery ML for Technical Debt
A case study utilizing BigQuery ML forecasting to predict operational engineering bottlenecks and tech debt.
-
BigQuery ML A/B Testing Use Cases
Apply statistical predictors to compare conversion rates and product options in BigQuery.
Prompt & Feature Engineering¶
Construct robust feature models and control model outputs:
-
:material-comment-text-prompt: GPT Prompts Reference
A compilation of structured prompt layouts, context stuffing rules, and formatting templates.
-
Prompt Engineering Guide
Master prompt strategies including few-shot, Chain-of-Thought, and self-evaluation loops.
-
Feature Engineering Fundamentals
Process variables, normalize values, and design input features for ML classifiers.
-
Feature Engineering & Observability
Align feature stores and runtime predictions with SRE metrics and drift telemetry.
AI Governance & Ethics¶
Manage organizational risk, fairness, and trust in AI systems:
-
AI Ethics & Governance
A leadership overview of compliance standards, safety policies, bias detection, and ethical alignment.