ŁѺ𝔚 Σ𐌍ŦȐΘℙΨ
Low Entropy Works explores how to reduce disorder in AI, products, and decision frameworks. The focus: clarity, ethical structure, and intelligence that behaves with intention.
What you'll find here:
- ① Product & AI strategy that scales.
- ② Responsible AI, governance and practical EU AI Act perspective.
- ③ Real product case studies: decisions, trade-offs, impact.
- ④ Applied data science and ML experiments.
- ⑤ Data-informed product growth & experimentation.
- ⑥ Frameworks for low-entropy decision making.
- ⑦ Product execution & delivery leadership.
- ⑧ Fractional product leadership.
A human behind:
Tatiana Bondarenko — technical Product Manager specialising in AI, data, and system clarity, building ethical, intelligent products.
Intersections:
Noise — Signal
Data — Meaning
Vision — Execution
Systems — People
Experimentation — Ethics
Velocity — Discipline
What I Do:
Product & Strategy
Create 0→1 products in AI, SaaS, and platform environments
Shape product vision and alignment
Design scalable AI-native systems
Compliance-aware product thinking (SOC2, ISO, EU AI Act)
Execution & Leadership
Lead cross-functional delivery
Drive growth through experimentation
Coordinate remote teams in complex environments
Technical & Data
Guide decisions with analytics
Hands-on with Python, SQL, Pandas, scikit-learn
Integrate APIs and design workflows
Collaborate on CI/CD, containerization, deployment
Now Exploring
- 2025-10Claim Fraud Detection (ML): Framing
- 2025-10Claim Fraud Detection (ML): Data Audit
- 2025-10Claim Fraud Detection (ML): Exploratory Analysis
- 2025-10Claim Fraud Detection (ML): Bias Surface Inspection
- 2025-10Claim Fraud Detection (ML): Feature Engineering
- 2025-10Claim Fraud Detection (ML): Leakage Prevention
- 2025-10Claim Fraud Detection (ML): Baseline Models
- 2025-10Claim Fraud Detection (ML): Benchmarking
- 2025-11Claim Fraud Detection (ML): Optimization
- 2025-11Claim Fraud Detection (ML): Threshold Tuning
- 2025-11Claim Fraud Detection (ML): Boosting Models
- 2025-12Claim Fraud Detection (ML): Interpretability (SHAP)
- 2025-12Claim Fraud Detection (ML): Bias & Fairness Check
- 2025-12Claim Fraud Detection (ML): Pipeline Consolidation
- 2025-12Claim Fraud Detection (ML): Streamlit Demo
Latest Work
Deep Dives
- 2025-01Ethical AI
- 2024-11Case Studies
- 2024-10Systems Frameworks
- 2024-10Signal vs Noise Taxonomy
- 2024-09AI Failure Modes
- 2024-09Cognitive Load in Product Decisions
- 2024-08Architectures for 0→1 Platforms
- 2024-08Boundary Objects in Cross-Functional Teams
- 2024-07Alignment Fractures in Scaling Organizations
- 2024-07Bias Gradients in Human-AI Interaction
Recommended Signals
Curated papers, posts, and research worth exploring.
- OpenAI Speculative Execution for Agents Parallelizing agent to reduce latency
- DeepMind Chain-of-Thought Safety Risks Long reasoning chains leak private data
- Amplitude Growth Feedback Loops Systems thinking for product expansion
- NYU Model Collapse in Fine-Tuning Explains why models degrade over time
- Meta FAIR Multimodal Evaluation Gaps Benchmarks behind generative capability