Smart-AI-Memory/empathy-framework

🎯 Transparent Cost Savings Analysis

This release replaces misleading blanket claims with honest, role-based cost savings estimates.

Key Changes

Tier Routing Savings: 34-86% depending on work role and task distribution

| Your Role | PREMIUM % | CAPABLE % | CHEAP % | Actual Savings |
|-----------|-----------|-----------|---------|----------------|
| Architect / Designer | 60% | 30% | 10% | 34% |
| Senior Developer | 25% | 50% | 25% | 65% |
| Mid-Level Developer | 15% | 60% | 25% | 73% |
| Junior Developer | 5% | 40% | 55% | 86% |
| QA Engineer | 10% | 35% | 55% | 80% |
| DevOps Engineer | 20% | 50% | 30% | 69% |

Key Insight: The often-cited "80% savings" assumes balanced task distribution (12.5% PREMIUM, 37.5% CAPABLE, 50% CHEAP). Architects and senior developers performing design work will see lower savings due to higher PREMIUM tier usage.

Documentation

• Role-Based Analysis
• Sensitivity Analysis
• Provider Comparison

Transparency Commitment

All claims backed by pricing math (Anthropic/OpenAI published rates) and task distribution estimates. No real telemetry data yet - v3.8.1 will add usage tracking for personalized savings reports.

Intelligent Caching System

• Hash-only cache: 30.3% average hit rate (100% on identical prompts)
• Hybrid cache: Up to 57% hit rate with semantic matching
• Benchmarked results: All claims backed by real benchmark data

See benchmark_semantic_cache.py for reproducible tests.

Bug Fixes

• Fixed exception handling specificity in config.py and refactor_plan.py
• Replaced broad except Exception: with specific exception types
• Added missing binascii import for proper error handling

What's Changed

• feat: v3.8.0 - Transparent Cost Claims & Role-Based Savings Analysis (bf2359a)
• docs: Add working principles to project memory (58e54dd)
• feat: Release v3.8.0 - Intelligent Caching with Benchmarked Results (3288e0b)
• fix: Add missing binascii import for proper error handling (0f7a4e1)
• fix: Replace broad exception handlers with specific types in long_term.py (1e70c28)
• fix: Resolve 4 critical type errors in cache module (37afc34)
• docs: Add comprehensive hybrid caching documentation (76a0da3)

Full Changelog: https://github.com/Smart-AI-Memory/empathy-framework/compare/v3.7.1...v3.8.0

Install: pip install empathy-framework==3.8.0