Purpose: Research-driven development methodology for technical decisions
- Complex technical decisions (architecture, algorithms, performance)
- Error fixes requiring deep understanding
- New feature implementation
- Performance optimization
1. Define Problem
- Specific technical challenge
- Success criteria and constraints
- Current limitations
2. Literature Review (Brave Search MCP)
- Academic papers and SOTA research
- Industry benchmarks and case studies
- Android-specific solutions
3. Technical Analysis (Context7 MCP)
- Android API capabilities
- Hardware constraints
- Integration complexity
4. Experimental Design
- Prototype plan
- Benchmarking approach
- Success metrics
5. Implementation & Validation
- Small proof of concept
- Performance measurement
- Documentation of findings
- Speech recognition models (Whisper variants, SOTA alternatives)
- Voice activity detection algorithms
- Model optimization (quantization, pruning)
- On-device ML performance
- FFmpeg Android integration
- Real-time audio pipelines
- Noise reduction algorithms
- Buffer management and threading
- Battery optimization
- Memory management
- CPU usage profiling
- Thermal management
- Audio storage and compression
- Privacy-first design
- Data lifecycle management
- Modular architecture patterns
- Research (Brave Search MCP + Context7 MCP)
- Document findings with sources
- Prototype small proof of concept
- Measure performance vs. research predictions
- Integrate incrementally with tests
- Validate against research targets
- Academic papers (peer-reviewed, recent)
- Industry documentation (Google, Android docs)
- Open source implementations (active projects)
- Community discussions (Stack Overflow, forums)
- Core Principles - Essential workflow rules
- Kotlin Migration Framework - Java-to-Kotlin conversion
- Performance Framework - Performance optimization
- Brave Search Guide - Research methodology
- Context7 Guide - Android documentation
- Research Template - Research documentation