This guide explores five practical estimation techniques—Fermi, analogy, expert judgment, Monte Carlo simulation, and three‑point PERT—detailing their principles, mathematical models, real‑world examples, and how to combine them for reliable decisions in business, engineering, and research.
This article explores how mathematical thinking—through concrete puzzles, probability paradoxes, and the Fermi estimation method—helps us overcome intuitive traps, make rigorous calculations, and apply logical reasoning to everyday problems and complex estimations.
This article explores the nature of agile estimation, details popular techniques such as Planning Poker, T‑Shirt Sizing, Story Points, Affinity and Wideband Delphi, and provides comprehensive risk‑management strategies and best‑practice recommendations to improve forecasting accuracy and project success.
This article reviews JD.com’s automation testing framework, explores how Conway’s Law and organizational culture affect software quality, presents a coupon‑system redesign case, and offers practical guidance on metrics, team cohesion, and probabilistic effort estimation for high‑quality delivery.
This article explains what story points are, how to estimate them using baseline stories and planning poker, compares them with person‑day estimates, outlines the benefits they bring to agile teams, and offers a practical workflow for adopting story‑point estimation.
The article reveals that professional developers spend most of their time maintaining large codebases, navigating domain knowledge, writing documentation, handling incompetent teammates, assuming bugs, estimating work, and attending meetings, emphasizing that delivering business value outweighs writing elegant code.
This article provides a detailed overview of user stories, their three essential elements, writing guidelines, role modeling, collection methods, acceptance testing, and practical application within Scrum, including estimation, prioritization, iteration planning, and the advantages and pitfalls of using user stories in software development.
Software engineers rarely build entire systems from scratch; instead they spend most of their time navigating large codebases, documenting, estimating, handling uncertainty, and collaborating with imperfect teammates, emphasizing business value over elegant code, and learning practical skills that university curricula often overlook.
This article reveals the hidden challenges of software engineering, emphasizing that most developers spend little time writing code and more on maintenance, documentation, dealing with incompetent colleagues, uncertainty, estimations, and meetings, urging newcomers to accept these realities and focus on delivering value.
This article explains why agile teams prefer story points over time for effort estimation, how the Fibonacci sequence (and its modified form) aligns with human perception of differences, the role of planning poker and the Delphi method in reducing bias, and practical steps for establishing baseline stories and estimating story points.
This article examines the most common obstacles agile testing teams encounter—from constantly changing requirements and neglecting cross‑browser testing to lack of automation, over‑emphasis on speed, unclear definitions of done, poor estimation, skill gaps, and technical debt—offering practical insights to address each issue.
This article demonstrates how to use Python to simulate coin tosses, construct the likelihood function for a Bernoulli distribution, and apply maximum likelihood estimation to recover the underlying head probability, illustrating the method with code, equations, and visualizations.
The article recounts how a major bank's R&D center used a card‑playing exercise to demonstrate agile story‑point estimation, detailing the step‑by‑step process, its purpose, and answering common questions about the technique’s applicability, units, and relative‑estimation logic.
This article explains how treating software projects as disciplined design processes—using shared language, realistic estimation, systematic system modeling, and test‑first development—helps create clean, maintainable code while avoiding the pitfalls of over‑engineering and rushed delivery.
This article explores the philosophy of clean code, emphasizing disciplined design, clear domain language, realistic estimation, collaborative system modeling, and test‑first development to produce maintainable, efficient software that meets client needs while avoiding unnecessary complexity.
This article introduces the most commonly used algorithms in engineering—recommendation, optimization, estimation, and classification—describes their typical application scenarios, and explores various testing methods and evaluation metrics such as offline experiments, user surveys, A/B testing, and performance indicators like accuracy, coverage, and robustness.
The article recounts a Nokia mobile‑software project where a newly appointed agile consultant analyses the company’s Scrum adoption, team structure, user‑story splitting, point‑based estimation, and the exclusion of testing and code‑review effort to improve planning and delivery for a ten‑month, 200‑person product development effort.
This article explains how a team transitions from waterfall to an Agile 101 approach by introducing time‑boxed iterations, applying the INVEST principle for user‑story splitting, using a ranking method for relative estimation, and outlining a practical planning process to improve project predictability.
