Skip to content

Examples Overview

Explore 6 real-world usage examples of OntoMem demonstrating different capabilities and use cases.

1️⃣ Self-Improving Debugger

An AI debugging agent that learns from each bug it encounters, building an evolving knowledge base of solutions.

Theme: Error Learning & Consolidation
Strategy: LLM.BALANCED merge with intelligent consolidation
Key Features: Error consolidation, LLM-powered merging, cross-encounter learning
View Example → | Source Code

2️⃣ RPG NPC Memory

Simulates an RPG game where NPCs build their memory of player characters through multiple interactions.

Theme: Character Profile Building
Strategy: MERGE_FIELD (incremental field updates)
Key Features: Multiple interaction types, progressive reputation tracking, incremental skill recognition
View Example → | Source Code

3️⃣ Semantic Scholar

Builds a searchable research paper library with semantic search capabilities. Papers can be discovered by content similarity, not just keywords.

Theme: Academic Paper Library
Strategy: Vector search + persistence
Key Features: Vector embeddings, semantic search (requires OpenAI API), metadata management
View Example → | Source Code

4️⃣ Multi-Source Fusion

Consolidates customer information from multiple systems (CRM, billing, support, marketing) into a unified profile using intelligent merging.

Theme: Customer Data Integration
Strategy: LLM.BALANCED merge with conflict resolution
Key Features: Multi-system integration, automatic conflict resolution, data quality reporting, lineage tracking
View Example → | Source Code

5️⃣ Conversation History

Shows how AI maintains and evolves its understanding of a user through multi-turn conversation.

Theme: Conversational Memory Evolution
Strategy: MERGE_FIELD with incremental fact accumulation
Key Features: Turn-by-turn updates, incremental fact accumulation, context maintenance
View Example → | Source Code

6️⃣ Temporal Memory Consolidation

Turn a stream of fragmented events into a single "Daily Summary" record using Composite Keys.

Theme: Time-Series Data Aggregation
Strategy: Time-Slicing with LLM.BALANCED merge
Key Features: Daily aggregation, temporal bucketing, time-aware consolidation
View Example → | Source Code

Running the Examples

All examples are included in the examples/ directory:

# Navigate to examples directory
cd examples/

# Run a specific example
python 01_self_improving_debugger.py
python 02_rpg_npc_memory.py
python 03_semantic_scholar.py
python 04_multi_source_fusion.py
python 05_conversation_history.py
python 06_temporal_memory_consolidation.py

# For Chinese versions
cd zh/
python 01_self_improving_debugger.py

Feature Matrix

# Example Theme Strategy Complexity API Required
01 Self-Improving Debugger Error Learning LLM.BALANCED ⭐⭐⭐ Optional
02 RPG NPC Memory Character Profiling MERGE_FIELD ⭐⭐ ❌ No
03 Semantic Scholar Research Library Vector Search ⭐⭐⭐ ✅ Yes
04 Multi-Source Fusion Data Integration LLM.BALANCED ⭐⭐⭐⭐ Optional
05 Conversation History Chat Memory MERGE_FIELD ⭐⭐⭐ ❌ No
06 Temporal Memory Time-Series LLM.BALANCED ⭐⭐⭐ Optional

Quick Start Examples

Check the Quick Start guide for immediate usage examples.

Next Steps