The Curious Engineer

All blogs / Distributed systems / cloud

Bytepawn

bytepawn.com

Distributed systems / cloudPythonAI / MLAlgorithms / data structures

’15’16’17’18’19’20’21’22’23’24’25’26
Posts
249 from 25 Jan 2015 to 11 Jun 2026
Posting
monthly (13 in the last year)
Kind
personal blog
Feed
https://bytepawn.com/feeds/all.atom.xml

2026 7

  1. The same server in five languages, and where they secretly disagree
  2. Writing a simple Go async message queue server
  3. András Lányi and the Problem of Human-Scale Politics
  4. Single round-trip optimization for Paxos with PaxosLease masters
  5. Combining Paxos and PaxosLease
  6. Writing a simple Rust async message queue server
  7. Multi-Paxos with Disk Persistence

2025 26

  1. What Ilya Sutskever, Satya Nadella, and Sam Altman think about AI progress in 2025
  2. What does ChatGPT think about Viktor Orban's 15 years as Hungary's leader?
  3. PaxosLease: Extending and Releasing Leases
  4. Implementing PaxosLease in Python with HTTP Flask
  5. Multi-Paxos: Building a Simple Replicated Log in Python
  6. Paxos: Lamport's Crown Jewel for Replicated State Machines
  7. Marketing Effectiveness and Experimentation Measurement
  8. Modern Monetary Theory: What’s Missing from the Narrative
  9. The MMT Case for Governments Funding Open Source Software
  10. A brief discussion of the Practical Byzantine Fault Tolerance (PBFT) algorithm
  11. Lamport's Byzantine Consensus algorithm with Signatures
  12. Applications of Byzantine Consensus
  13. A more general implementation of Lamport's Byzantine Consensus algorithm in Python
  14. Newer Posts
  15. Lamport's Byzantine Generals algorithm in Python
  16. Lamport's logical clocks and the Ricart–Agrawala Distributed Mutual Exclusion algorithm in Python
  17. Exploring Lamport's Bakery algorithm in Python
  18. Exploring Lamport's Bakery algorithm in C++20
  19. Karpathy: AI is changing Software Engineering
  20. Writing a simple async message queue server in C++, Python and JavaScript — Summary
  21. Writing a simple C++20 async message queue server - Part III
  22. White noise into random walks
  23. Random Walks and the Dickey-Fuller Test - Part II
  24. Random Walks and the Dickey-Fuller Test - Part I
  25. The reward for long-term resilience in simple trading strategies - Part II
  26. The reward for long-term resilience in simple trading strategies - Part I

2024 42

  1. Culture talk: What I learned from building a distributed database that never took off
  2. Modern Portfolio Theory III: The Efficient Frontier in log volatility—return space
  3. The MMT Perspective on Crypto
  4. I ❤️ triathlons
  5. Modern Portfolio Theory II: Random portfolio coverage in log volatility—return space with a risk free asset
  6. Individual Career Plan
  7. Why I read history books
  8. Paper: The Unreasonable Effectiveness of Monte Carlo Simulations in A/B Testing
  9. Older Posts
  10. Modern Portfolio Theory I: Random portfolio coverage in log volatility—return space
  11. The Unreasonable Effectiveness of Monte Carlo Simulations in A/B Testing
  12. A simple telnet based Taboo game in async Python
  13. AI Strategy in 2024 for the Rest of Us
  14. Techtalk: Why are A/B tests the gold standard of causal inference?
  15. 200 articles
  16. ChatGPT's critique of OpenAI CEO Sam Altman's vision of the AI future
  17. How I manage my 1v1s
  18. Breaking Bell's Inequality with Monte Carlo Simulations in Python
  19. Quantum Entanglement, Locality, Determinism
  20. Why I don't live in Hungary
  21. How to solve it
  22. How I screen CVs
  23. Writing a simple Javascript async message queue server - Part III
  24. Writing unit tests for the async message queue server
  25. Testing randomness extractors
  26. Randomness extractors: making fair coins out of biased coins
  27. Writing a simple Python async message queue server - Part III
  28. Writing a simple Javascript async message queue server - Part II
  29. Writing a simple Javascript async message queue server - Part I
  30. Older Posts
  31. Paper review: A Comparison of Approaches to Advertising Measurement
  32. The Culture of Powerpoint
  33. Student paper: The Exciting Physics of an Excited Universe
  34. Paper: Monte Carlo Experiments of Network Effects in Randomized Controlled Trials
  35. Peak personal computing, or, my current workstation setup
  36. I ❤️ open source, and, Cloud9 still rocks
  37. Papers
  38. Stock returns of simple trading strategies
  39. The volatility of simple trading strategy returns
  40. Introductory investigations into the stability of stock price volatility
  41. 2024 Data Outlook
  42. How to achieve high performance, and, ratings

2023 25

  1. Wordcount III: Beating the system `wc`
  2. Wordcount II: Introducing a cleaner C++ class hierarchy to `wc`
  3. Wordcount I: Implementing the Unix command-line tool `wc` in modern C++
  4. Better be first 99% of the time than second 100% of the time
  5. Introduction to Marketing Mix Modeling
  6. Leadership models V: The Hero's Journey
  7. Real-world experiments: 5 Lessons from Google, Bing, Netflix and Alibaba
  8. Conditional Probabilities and Simpson's Paradox
  9. Testing GPT-4 spatial reasoning and comprehension
  10. Leadership models IV: the Pareto Principle, the Peter Principle, the Rumsfeld Matrix, Servant Leadership and Pygmalion Effect, and Goleman's Emotional Intelligence Model
  11. Leadership models III: First Principles Thinking
  12. Leadership models II: Growth Mindset, Eisenhower Matrix, Tuckman Model, Cynefin Framework, SCARF Model
  13. Leadership models I: Iceberg Model, Six Thinking Hats, Trust Equation, Circle of influence, OODA Loop
  14. GPT News Poet: silly AI poems based on today's news
  15. Can ChatGPT write the C++20 async message queue codes from the previous posts?
  16. Writing a simple C++20 async message queue server - Part II
  17. Writing a simple C++20 async message queue server - Part I
  18. Can ChatGPT write the Python async message queue codes from the previous posts?
  19. Writing a simple Python async message queue server - Part II
  20. Writing a simple Python async message queue server - Part I
  21. Lifelong purchases
  22. Crypto V: On NFTs
  23. Crypto IV: On stock investing vs. crypto investing
  24. Crypto III: On perceivd crypto advantages such as anonymity and irreversibility
  25. Crypto II: On fiat currencies vs. crypto currencies

2022 40

  1. Crypto I: On cryptocurrencies, explained using FTX Tokens
  2. Finding similar API functions between Pytorch and Tensorflow with Doc2Vec
  3. Testing GPT-3 spatial reasoning and comprehension
  4. Fred Brooks' The Mythical Man-Month
  5. How good an astrophysicist is GPT-3?
  6. Similar posts recommendation with Doc2Vec - Part III
  7. Similar posts recommendation with Doc2Vec - Part II
  8. Similar posts recommendation with Doc2Vec - Part I
  9. Ask HN: Data Scientists, what libraries do you use for timeseries forecasting?
  10. Estimating mathematical constants with Monte Carlo simulations
  11. Common patterns in technical interviewing
  12. How good a particle physicist is GPT-3?
  13. Data Science Culture Doc
  14. How good a Data Scientist is GPT-3? - Part II
  15. How good a Data Scientist is GPT-3?
  16. The Elements of Typographic Style
  17. The Culture Map
  18. Working Backwards
  19. HBR Guide to Better Business Writing
  20. Culture Docs: Facebook, Netflix and Valve
  21. More Data Scientists should learn SQL
  22. Useful Python decorators for Data Scientists
  23. Building a toy Python @dataclass decorator
  24. Python decorator patterns
  25. Building a toy Python Enum class - Part II
  26. Building a toy Python Enum class - Part I
  27. "Over 70% of all Porsche vehicles ever built are still on the road today"
  28. "The company is all hot air"
  29. Python types for Data Scientists - Part III
  30. Python types for Data Scientists - Part II
  31. Python types for Data Scientists - Part I
  32. Solving 5 algorithmic interview questions
  33. Fair coin from biased coin
  34. The german tank problem in World War II
  35. Delightfully cynical half-truths about organizations
  36. Probabilistic spin glass - Conclusion
  37. Optimal coverage for Wordle with Monte Carlo methods - Part III
  38. Optimal coverage for Wordle with Monte Carlo methods - Part II
  39. Optimal coverage for Wordle with Monte Carlo methods - Part I
  40. Probabilistic spin glass - Part V

2021 40

  1. Probabilistic spin glass - Part IV
  2. Probabilistic spin glass - Part III
  3. Probabilistic spin glass - Part II
  4. Probabilistic spin glass - Part I
  5. The physical Sackur-Tetrode entropy of an ideal gas
  6. Entropy of an ideal gas with coarse-graining
  7. WeToddle retrospective
  8. Entropy in Data Science
  9. 100 articles
  10. Cross entropy, joint entropy, conditional entropy and relative entropy
  11. What's the entropy of a fair coin toss?
  12. Five ways to reduce variance in A/B testing
  13. Correlations, seasonality, lift and CUPED
  14. A/A testing and false positives with CUPED
  15. Reducing variance in conversion A/B testing with CUPED
  16. Reducing variance in A/B testing with CUPED
  17. A/B testing and the historic lift paradox [redacted]
  18. The Full Stack Data Scientist
  19. Comparing NeuralProphet and Prophet for timeseries forecasting
  20. Timeseries forecasting with Prophet
  21. YOLO object detection architecture
  22. YOLOv5 object detection experiments
  23. Predicting party affiliation of US politicians using fasttext
  24. Random digits and Benford's law
  25. Tricks vs implementation in coding interviews
  26. 10 ways to iterate from 0 to 1 with deciles
  27. Sometimes brute forcing just works
  28. Building intuition for p-values and statistical significance
  29. Random numbers, the natural logarithm and higher dimensional simplexes
  30. Classification accuracy of quantized Autoencoders with Pytorch and MNIST
  31. Investigating information storage in quantized Autoencoders with Pytorch and MNIST
  32. Building a Pytorch Autoencoder for MNIST digits
  33. The best parts of Invent and Wander: the Collected Writings of Jeff Bezos
  34. Training a Pytorch Wasserstein MNIST GAN on Google Colab
  35. Training a Pytorch Classic MNIST GAN on Google Colab
  36. How I manage notifications to keep myself uninterrupted
  37. Training a Pytorch Lightning MNIST GAN on Google Colab
  38. Automatic MLFlow logging for Pytorch
  39. Automatic MLFlow logging for Scikit Learn
  40. Getting Started with MLFlow

2020 26

  1. Making statistics lie for the 2020 Presidential election
  2. Comparing conversion at control and treatment sites
  3. Unevenness at the edges
  4. Effective Data Visualization Part 3: Line charts and stacked area charts
  5. Effective Data Visualization Part 2: Formatting numbers
  6. Effective Data Visualization Part 1: Categorical data
  7. Multi-armed bandits and false positives
  8. A/B testing and Multi-armed bandits
  9. Understanding Facebook’s Planout A/B testing framework
  10. Validation checks for A/B tests
  11. Running multiple A/B tests in parallel
  12. Bayesian A/B conversion tests
  13. A/B testing and the G-test
  14. A/B testing and networks effects
  15. A/B testing on social networks
  16. Early stopping in A/B testing
  17. A/B testing and Fisher's exact test
  18. A/B testing and the Chi-squared test
  19. A/B testing and the t-test
  20. A/B testing and the Z-test
  21. Beyond the Central Limit Theorem
  22. A/B testing and the Central Limit Theorem
  23. Optimizing waits in Airflow
  24. SQL best practices for Data Scientists and Analysts
  25. How I write SQL code
  26. Small team planning

2019 24

  1. Personal goaling
  2. Pytorch in 2019
  3. Warren Buffett style fundamental metrics of long-term company performance
  4. Calibration curves for delivery prediction with Scikit-Learn
  5. Using simulated self-play to solve all OpenAI Gym classic control problems with Pytorch
  6. Applying policy gradient to OpenAI Gym classic control problems with Pytorch
  7. Machine Learning at Fetchr
  8. Solving the CartPole Reinforcement Learning problem with Pytorch
  9. Metrics Atlas
  10. Playing Go with supervised learning in Pytorch
  11. Arabic name classification with Scikit-Learn and Pytorch
  12. Exploring prior beliefs with MCMC
  13. A/B tests: Moving Fast vs Being Sure
  14. Food deliveries, Bayes and Computational Statistics
  15. The Collatz conjecture
  16. MNIST pixel attacks with Pytorch
  17. Solving CIFAR-10 with Pytorch and SKL
  18. Solving MNIST with Pytorch and SKL
  19. SVM with Pytorch
  20. Hacker News Embeddings with PyTorch
  21. rxe: literate and composable regular expressions
  22. PyTorch Basics: Solving the Ax=b matrix equation with gradient descent
  23. Automating a Call Center with Machine Learning
  24. 5 things that happened in Data Science in 2018

2018 6

  1. Warehouse locations with k-means
  2. Growth Accounting and Backtraced Growth Accounting
  3. Fetchr Data Science Infra at 1 year
  4. What not to spend time on
  5. Beat the averages
  6. Building the Fetchr Data Science Infra on AWS with Presto and Airflow

2016 11

  1. Don’t build cockpits, become a coach
  2. Beautiful A/B testing
  3. Hack, HHVM and avoiding the Second-system effect
  4. Einstein's amazing theory
  5. Heisengames and the importance of patience in business
  6. Cloud9: Cloud coding that actually works
  7. Luigi vs Airflow vs Pinball
  8. Pinball review
  9. How to make a blog like this
  10. Airflow review
  11. Systems thinking and system traps

2015 2

  1. Luigi review
  2. Cargo Cult Data