GV

GeezerVison

The World Wide Web, restored to 800 pixels of glory.

Home|About|Gaming cabinet|Archives|How it works|Starter sites

On this site

No menu could be recovered. Follow the blue links in the page.

This view

  • Original page
  • New address

GeezerVison is showing https-deeplearning-ai.github.io — Context Engineering Course — Interactive Trajectory Visuals

Module 1

Failure Modes in Multi-Turn Agents

Context Poisoning

Trajectory 1: Customer Service Agent

An agent handling shoe size exchanges ingests unindexed, conflicting legacy documentation (30-day cutoff), poisoning its reasoning and wrongly rejecting an eligible request under the 60-day policy.

Launch Visual App
Rapid Saturation

Trajectory 2: Customer Service Agent

Multi-hop logistics queries for regional inventory stock and freight dispatch return large, unpruned payloads, causing explosive context growth and saturating the token budget.

Launch Visual App
Module 2

Agentic Search & Retrieval

Strategy Mismatch

Trajectory 1: FreshCart Coding Agent

The agent attempts to locate exact code constants using pure vector semantic search and policy concepts using keyword search, failing to adapt or recover without hybrid retrieval.

Launch Visual App
Naive Chunking

Trajectory 2: FreshCart Coding Agent

Fixed character-count splitting without language-aware AST parsing splits model validation logic across boundaries, leading to hallucinated data constraints.

Launch Visual App
Module 3

Context Formatting & Compaction

Compaction Erosion

Trajectory 1: Cinema Research Agent

Compacting summaries of summaries across turns progressively strips vital clues (jazz score, European jury award) needed to identify an obscure animated film.

Launch Visual App
Missing Pointer

Trajectory 2: Cinema Research Agent

Externalizing full document bodies to a scratchpad without returning the external document ID pointer strands retrieved records outside the agent's reach.

Launch Visual App
Module 4

Agent Memory & State Persistence

Memory Deduplication

Trajectory 1: Doctor's Appointment Assistant

Demonstrates memory writing pipelines reconciling newly paraphrased patient preferences against long-term vector memory to prevent duplicate rule entries.

Launch Visual App
Forgotten Rules

Trajectory 2: Doctor's Appointment Assistant

A stateless assistant using exact flat-file keyword search fails to equate "Friday" with "end-of-week", forgetting user preferences across sessions.

Launch Visual App
You are visitor 01402520 — GeezerVison does not store the pages you look up.
Best viewed with Netscape Navigator 4.7 or Internet Explorer 6 at 800×600. No cookies required.