Skip to content
PodParley
AI Intuition podcast artwork

PODCAST · technology

AI Intuition

by Dan Sarmiento

This is the gold rush era of artificial intelligence. You want to learn quickly so you don't get left behind, but how can you learn about AI without an advanced degree in computer science and mathematics?You translate all the complicated concepts into plain language and you summarize the relevant news into a podcast you can listen to while you do everything else.This is the method that helped me speed up my learning and maybe it can help you too.

Publisher-supplied feed metadata · PodParley refreshed Sep 3, 2026 · Source feed

Subscribe · 0 Bookmark

Episodes

  1. Google Agentic Case Study - Episode 6: Continuous Evaluation, Shadow Operations & Closed-Loop MLOps

    This episode highlights the "day-2" operational pipeline: pre-deployment evaluation gates, shadow-mode testing, demographic fairness monitoring, and the continuous learning flywheel.

  2. Google Agentic Case Study - Episode 5: Infrastructure Deployment & Security Validation

    This episode details how to build and secure a HIPAA-compliant Healthcare Landing Zone using modern cloud security frameworks and automated infrastructure tools.

  3. Google Agentic Case Study - Episode 4: Safety Guardrails, Blast-Radius Control & Fault Tolerance

    This episode covers the crucial safety mechanisms, deterministic validation gates, de-identification pipelines, and fallback strategies that protect clinical systems from failure.

  4. Google Agentic Case Study - Episode 3: Component Prototyping & Technical Implementation

    This episode focuses on isolating individual microservices, prompt engineering system instructions, and leveraging concurrent execution to keep latency to a minimum.

  5. Google Agentic Case Study - Episode 2: Cognitive Architecture & Multi-Agent Design

    This episode explores how to coordinate specialized AI agents, design multi-tier memory systems, and manage active session states without data race conditions.

  6. Google Agentic Case Study - Episode 1: Requirements Scoping, Regulatory Boundaries & Clinical Benchmarking

    This episode focuses on defining the administrative problem space, curating the "golden dataset," establishing clinical rules, and aligning with national healthcare interoperability mandates.

  7. Google Agentic Architect - Subconcept 5.2: Distributed Trace & Telemetry

    Deep dive into post-deployment monitoring and tracing for complex, multi-turn, multi-agent systems. Explain how to instrument agent code with OpenTelemetry GenAI semantic conventions (tracking span attributes like model, tokens, and temperatures). Detail how standard W3C traceparent headers must...

  8. Google Agentic Architect - Subconcept 5.1: Pre-Production Continuous Evaluation Gates (CI/CD)

    Detail how to construct automated pre-production evaluation pipelines for non-deterministic agents. Walk through building an evaluation harness (e.g., tests/eval_agent_trajectory.py using pytest and Cloud Build) executed against static "golden datasets." Contrast three evaluation types: Pointwise...

  9. Google Agentic Architect - Subconcept 4.3: Deterministic Tool Control

    Explain how to eliminate non-deterministic behavior when models interact with transactional enterprise systems. Detail the strict parameter tunings required: setting temperature=0.0, enabling greedy decoding (top_k=1), and enforcing strict JSON schemas on model outputs. Explain how to isolate...

  10. Google Agentic Architect - Subconcept 4.2: Semantic Threat Mitigation

    Analyze the runtime semantic firewalls required to protect models from real-time exploits. Focus on the integration of Vertex AI Model Armor as an inline security gateway to detect and block prompt injections, jailbreaks, and sensitive data leaks (configuring the filter confidence level to...

  11. Google Agentic Architect - Subconcept 4.1: Cryptographic Network Perimeters & Access Control

    Detail the zero-trust infrastructure required to secure enterprise agent runtimes. Explain how to establish VPC Service Controls (VPC-SC) to create cryptographic network perimeters enclosing models, databases, and buckets to prevent data exfiltration. Discuss the Principle of Least Privilege...

  12. Google Agentic Architect - Subconcept 3.3: Context Window Optimization & Cost Controls

    Address the token economics and operational latency of long-running agent sessions. Detail three key optimizations: 1) Sliding Window Pruning, explaining how to maintain a rolling history of only the last K=6 to 10 active turns while archiving older interactions; 2) Asynchronous Map-Reduce...

  13. Google Agentic Architect - Subconcept 3.2: Concurrent State Mutation Safety

    Focus exclusively on the critical concurrency challenges in multi-agent architectures where parallel worker nodes attempt to update a shared session state. Deep dive into the implementation of Optimistic Concurrency Control (OCC) using Cloud Firestore update preconditions. Explain how to design...

  14. Google Agentic Architect - Subconcept 3.1: Four-Tier Memory Architecture

    Deconstruct the physical storage backends and latency profiles of the four-tier agent memory hierarchy. Explain the operational characteristics of: 1) Short-Term Working State (Python in-memory heap or Memorystore/Redis, sub-5ms), 2) Durable Session State (Cloud Firestore or…

    more

  15. Google Agentic Architect - Subconcept 2.3: Agent Skills Packaging

    Detail the modern, modular developer lifecycle for building and deploying "Agent Skills." Explain how skills package behavioral instructions, tool schemas, and dependency manifests (such as skill.yaml files) into immutable, versioned OCI artifacts distributed via Google Cloud Artifact Registry....

  16. Google Agentic Architect - Subconcept 2.2: Tool Augmentation & Grounding

    Analyze the technical implementation of tool augmentation and grounding within the Gemini and Vertex AI ecosystem. Break down Vertex AI Search managed grounding, focusing on how document ingestion, chunking, and semantic retrieval produce explicit citations ("grounding_chunks") to eliminate model...

  17. Google Agentic Architect - Subconcept 2.1: Managed Runtimes & Compute Platforms

    Compare and contrast the three primary runtimes for hosting enterprise agent workloads on Google Cloud: Vertex AI Agent Engine (Reasoning Engine), Cloud Run, and GKE Autopilot. Focus intensely on the serialization mechanics of Vertex AI Agent Engine: explain how custom Python agent classes are...

  18. Google Agentic Architect - Subconcept 1.3: Inter-Agent Protocols & Context Slicing

    Detail how multi-agent systems communicate and maintain context sanity without flooding context windows. Deep dive into the Agent-to-Agent (A2A) protocol using typed, secure Pydantic message envelopes (like A2AMessageEnvelope). Explain the Model Context Protocol (MCP) as a standardized,...

  19. Google Agentic Architect - Subconcept 1.2: Multi-Agent Topologies (The Coordination Mesh)

    Contrast three primary multi-agent orchestration topologies: Router (Triage/Dispatch), Hierarchical Supervisor-Worker, and Sequential DAG pipelines. Explain the operational trade-offs of each pattern regarding orchestration overhead, end-to-end latency, and state complexity. Describe a router as…

    more

  20. Google Agentic Architect - Subconcept 1.1: Cognitive Reasoning Loops (The "Brain" of the Agent)

    Conduct an advanced technical deep dive into the inner workings of three cognitive reasoning loops: ReAct (Reasoning + Acting), Plan-and-Solve (DAG-based), and Self-Reflective (Actor-Critic) refinement. Explain the exact mechanics of a ReAct loop: how the model generates an internal Thought…

    more

  21. Google Antigravity Mastery Series - Subconcept 6.2: Python Determinism Libraries (Under-the-Hood Engineering)

    Deep-diving into the programmatic libraries developers use to enforce syntactically correct outputs, validation pipelines, and strict state management.

  22. Google Antigravity Mastery Series - Subconcept 6.1: Core Cognitive Execution Models

    Comparing single-threaded reasoning models with advanced tree-search and reinforcement learning systems, detailing their execution trade-offs and error-propagation vulnerabilities.

  23. Google Antigravity Mastery Series - Subconcept 5.3: Model Context Protocol (MCP) Integration

    Connecting the terminal agent to external tools, databases, and collaboration platforms using standardized, model-agnostic transports.

  24. Google Antigravity Mastery Series - Subconcept 5.2: Lifecycle Event Hooks

    Implementing enterprise guardrails, automated code formatting, and security policies by intercepting execution at key boundaries.

  25. Google Antigravity Mastery Series - Subconcept 5.1: The Antigravity Python SDK (google.antigravity)

    Scripting autonomous behavior programmatically, defining Pydantic schemas for structured JSON output, and running sandboxed custom execution blocks.

  26. Google Antigravity Mastery Series - Subconcept 4.2: Team Communication Topologies

    Architectural trade-offs between Hierarchical, Horizontal, and Blackboard multi-agent systems, and how they manage context boundaries.

  27. Google Antigravity Mastery Series - Subconcept 4.1: Background Agents & Non-Blocking TUI Control

    Managing background execution threads, using the /agents panel control room, and keeping active context clean using side-channels.

  28. Google Antigravity Mastery Series - Subconcept 3.3: Multimodal & Target Grounding

    Bypassing noisy vector search search-hallucinations via precise @ targeting and feeding screenshots for UI testing

  29. Google Antigravity Mastery Series - Subconcept 3.2: Modular Skill Frameworks (SKILL.md)

    Structuring reusable, on-demand domain playbooks using progressive disclosure to keep active context buffers low and clean

  30. Google Antigravity Mastery Series - Subconcept 3.1: Project Guideline Anchoring (GEMINI.md / AGENTS.md)

    Establishing permanent repository-level rules to enforce compliance, restrict forbidden packages, and combat instruction drift in long sessions

  31. Google Antigravity Mastery Series - Subconcept 2.4: Terminal TUI Navigation & Slash Commands

    Driving the terminal control center like an absolute power-user using keyboard hotkeys and specialized slash systems.

  32. Google Antigravity Mastery Series - Subconcept 2.3: Workspace Isolation & Rollbacks

    Managing git state and speculative filesystems to run highly destructive refactors with zero risk.

  33. Google Antigravity Mastery Series - Subconcept 2.2: CLI Configuration Parameters & Tuning

    Mastering runtime configuration flags to optimize performance, containment, reasoning depth, and context density.

  34. Google Antigravity Mastery Series - Subconcept 2.1: The OPAV (Observe-Plan-Act-Verify) Loop

    Understanding the internal step-by-step state machine of how the CLI processes a command from start to finish.

  35. Google Antigravity Mastery Series - Subconcept 1.3: Loop Bounds & Circuit Breakers

    Preventing runaway recursive execution loops, enforcing safety verification gates, and utilizing shadow rollbacks

  36. Google Antigravity Mastery Series - Subconcept 1.2: Context Window Dynamics & Decay Management

    Managing finite cognitive budgets, combatting token degradation, and using persistent files to prevent instruction drift

  37. Google Antigravity Mastery Series - Subconcept 1.1: Intent Articulation & Architectural Invariants

    Shifting from imperative, procedural commands to declaring system goal states, setting operational boundaries, and using adversarial alignment tools

  38. Microsoft Data Agent Development in Fabric

    An introduction to Microsoft Fabric data agents, which are conversational Q&A systems powered by generative AI that allow users to retrieve insights from governed OneLake data using natural language. It provides comprehensive instructions on configuring, validating, and managing these agents...

  39. Agent Builder by Docker

    cagent, Docker's open-source, multi-agent runtime designed to orchestrate autonomous AI systems by allowing users to build and manage teams of specialized AI agents. cagent uses a declarative YAML configuration for defining agents and their interactions, with a hierarchical structure where a root...

  40. Open Agentic Web Development - Project NANDA (MIT)

    Project NANDA, an initiative by the MIT Media Lab aimed at creating the foundational infrastructure for the "Open Agentic Web," an internet designed for autonomous AI agents rather than human users. This new architecture addresses the limitations of the current internet for agent discovery,...

  41. AI Startup Failure Analysis

    examines the paradox of unprecedented investment in the artificial intelligence sector coexisting with an accelerating rate of startup failures. It identifies a failure rate exceeding 90% for AI startups, significantly higher than the broader tech industry. The analysis categorizes these failures...

  42. AI Security - Model Denial of Service

    Model Denial of Service (Model DoS) attacks, a modern evolution of traditional DoS that targets the computational resources of AI and Machine Learning systems, rather than network bandwidth. It explains how these attacks degrade performance or render AI models unavailable, often by exploiting…

    more

  43. AI Security - Training Data Attacks

    analysis of training data poisoning, a critical integrity attack against AI and ML systems. It explains how adversaries corrupt the foundational learning phase by manipulating datasets, leading to altered model behavior, ranging from performance degradation to hidden backdoor attacks. The text...

  44. AI Security - Insecure Output Handling

    analysis of Insecure Output Handling, a critical application security vulnerability distinct from insecure input handling, emphasizing the need to never trust data sent to an interpreter. It details the diverse and severe consequences of this flaw, including client-side attacks like Cross-Site...

  45. AI Security - Prompt Injection

    analysis of prompt injection, which is identified as the leading security vulnerability in applications powered by Large Language Models (LLMs). It explains that this threat arises from the inherent architecture of LLMs, which struggle to differentiate between trusted developer instructions and...

  46. Unsupervised ML for Test Suite Reduction - Test Smarter Not Harder

    This research systematically maps literature concerning the application of unsupervised machine learning approaches to test suite reduction (TSR), a critical process for optimizing software testing efficiency. The study, which reviewed 34 papers published between 2013 and 2023, identifies common...

  47. bytedance USO - Unified Style and Subject-Driven Generation via Disentangled and Reward Learning (Image Model)

    analyze USO, a novel generative AI model developed by Bytedance's Intelligent Creation Lab. USO addresses the long-standing challenge of separately controlling style and subject in image generation by proposing a unified framework that synergizes these tasks. The text details USO's conceptual...

  48. Supervised Fine-Tuning on OpenAI Models

    overview of Supervised Fine-Tuning (SFT) for large language models, explaining it as a method to specialize pre-trained models for particular tasks by training them on curated, labeled datasets. It compares full fine-tuning with more efficient Parameter-Efficient Fine-Tuning (PEFT) methods like...

  49. NVIDIA's Jet Nemotron - Post Neural Architecture Search & JetBlock

    NVIDIA's new Jet-Nemotron model family, which introduces a hybrid-architecture approach to Large Language Models (LLMs) to significantly improve efficiency without sacrificing accuracy. This innovation is primarily driven by two key technologies: Post Neural Architecture Search (PostNAS), a…

    more

  50. What is Nano Banana - Google's Viral Image Generation Model

    Google's Gemini 2.5 Flash Image model, initially known by its codename "nano banana," highlighting its unconventional market entry through anonymous competitive testing on LMArena, which generated significant community-driven hype. The text explains its natively multimodal architecture…

    more
View all 127 episodes →

Type above to search every episode's transcript for a word or phrase. Matches are scoped to this podcast.

Searching…

We're indexing this podcast's transcripts for the first time — this can take a minute or two. We'll show results as soon as they're ready.

No matches for "" in this podcast's transcripts.

Showing of matches

No topics indexed yet for this podcast.

Loading reviews...

ABOUT THIS SHOW

This is the gold rush era of artificial intelligence. You want to learn quickly so you don't get left behind, but how can you learn about AI without an advanced degree in computer science and mathematics?You translate all the complicated concepts into plain language and you summarize the relevant news into a podcast you can listen to while you do everything else.This is the method that helped me speed up my learning and maybe it can help you too.

HOSTED BY

Dan Sarmiento

CATEGORIES

technology

Frequently Asked Questions

How many episodes does AI Intuition have?

AI Intuition currently has 127 episodes available on PodParley. New episodes are automatically indexed when they're published to the podcast feed.

What is AI Intuition about?

This is the gold rush era of artificial intelligence. You want to learn quickly so you don't get left behind, but how can you learn about AI without an advanced degree in computer science and mathematics?You translate all the complicated concepts into plain language and you summarize the relevant...

How often does AI Intuition release new episodes?

AI Intuition publishes regularly with 127 episodes available. Check the episode list for the typical release schedule.

Where can I listen to AI Intuition?

You can listen to AI Intuition on PodParley by clicking any episode. We provide an embedded audio player for direct listening, and you can also subscribe via your preferred podcast app using the RSS feed.

Who hosts AI Intuition?

AI Intuition is created and hosted by Dan Sarmiento.
URL copied to clipboard!