Vous êtes déchiré entre le respect des délais et le perfectionnement de la conception logicielle. Comment trouver le bon équilibre ?

There are several ways to effectively prioritize. - Segregate tasks into 2 buckets-one that map to Key Objectives for a given function and second which are not directly contributing to that goal. - Leverage Urgent vs Important Eisenhower framework - Classify tasks into P0- Enterprise Impacting, P1- Function impacting, P2- Limited Users Impacting and define timelines according to impact by each. - Classify based on Business value (in terms of metrics that are meaningful for industry an organization operates in), complexity to execute, user impact a feature may have. - Classify based on Non-discretionary (mandatory, Must-do), vs Good to have (can wait, plan these and execute with support).

No es posible perfeccionar completamente un diseño ya que un buen diseño se va refinando con el tiempo. Se debe respetar los plazos es fundamental para no quedar atrapado en un ciclo interminable sin entregas. Deben haber unos objetivos a la hora de entregar, entre ellos, deberían estar: Satisfacer los requisitos y especificaciones técnicas, cumplir con los estándares de calidad y completar las pruebas de UAT y completarse dentro del plazo acordado. Si por lo que fuese, esto no pudisese lograrse se debería pedir una extensión de plazo y añado que es importante hablar con las partes interesadas para decidir si prefieren: Extender el plazo o bien planificar una nueva versión (Si se detecta que el código puede mejorarse).

Balancing deadlines and software design requires prioritizing critical features, adopting iterative development practices, and managing time effectively. Use agile methodologies, conduct regular reviews, and set realistic deadlines. Implement automated testing and continuous integration to ensure quality. Plan for refactoring to manage technical debt, and use code quality tools to maintain standards. Communicate transparently with stakeholders and engage in continuous learning through post-mortems and professional development. This structured approach ensures the delivery of valuable, high-quality software within set timelines.

Priorizar o desenvolvimento iterativo é mais vantajoso do que esperar pela perfeição antes de lançar um produto. Especialmente em produtos que podem causar um grande impacto quando lançados rapidamente, é crucial adotar a mentalidade de "lançar rapidamente, aperfeiçoar depois". No entanto, o mais importante é ter um bom planejamento.

I have been put in this positions so many times, and what's more presurring having to do this across multiple projects!!, here are a few tips that helped me throughout the whole things : 1- if you got the chance always try to be clear about the scope of your work in the project and on your way of executing that work and communicate that to the stakeholders and make them commit to it!.

Working in sprints is great for decomposing the work and delivering incrementally. But in most cases I've found that there is still a deadline for delivery that doesn't care about your sprint cadence. That's when other compromises have to be made. Getting the hard work done in early sprints (easier said than done) can refund this risk.

break down large tasks into smaller - not more than a sprint worth of tickets , differentiate between investigative tickets vs implementation, demo and repeat

Focus on delivering stakeholder value incrementally. It’s not just about delivering something each sprint, but making sure that what you’re delivering each sprint is compelling to one or more stakeholders. Is it helping your product owner to hone their vision or become more confident about what they’re building? Is it giving the users something compelling that they can start using, even if it is not the final project deliverable? Is it giving the development team a strong foundation for confidence about how they’re building the features? And talk about this stakeholder value orientation each sprint to help everybody think of their work from this perspective. It can improve contributors’ motivation and satisfaction.

Agile approach definitely provides a robust framework for iterative development and frequent reassessment of priorities but this sometimes leads to compromise on deadlines because agile principal for software development does support adjustments in design as and when discovered. The complete software design is not thought through to start the development and as a result these changes are anticipated to deliver quality software product. Another key aspect of software designing to embrace Agile approach is Vertical slicing, but this also needs lot of efforts to divide the software capabilities in parts which can be delivered independently. Having buffer time to accommodate changes and design the software with smaller and easy to adjust blocks

Iterative growth is better than delayed perfection. That's the rule everyone should follow, especially for products that can impact more the sooner they launch. Ship fast, perfect later, is the mantra. For products, where perfection is absolutely necessary and mission critical such as cyber security products, deadlines must be cogent.

When everyone understands the current status and challenges, it becomes easier to prioritize tasks. Stakeholders can collectively decide which features are essential for the initial release and which can be deferred or improved later. This ensures that efforts are focused on delivering the most critical functionalities first.

Just always keep in mind that the first draft and the v.1.0 of the project will never have the perfect design,sometimes stakeholders cares more about delivery time more than design .

La comunicación es la columna vertebral de un proyecto. Integra la jerarquía equipo/cliente/empresa a nivel vertical y la coordinación del equipo a nivel horizontal. Si no hay comunicación efectiva y una estrategia, el proyecto esta condenado a sufrir problemas de gestión y retrasos.

Se debe aplicar principios de Comunicación Efectiva, esto fomenta un ambiente de colaboración, transparencia y confianza, lo que ayuda a encontrar el equilibrio adecuado entre cumplir plazos y perfeccionar diseño; no debemos dejar de un lado aplicar canales de comunicación. En mi caso particular me gusta mucho reuniones diarias (daily stand-ups), que sean cortas y concretas y por supuesto reuniones de revisión de sprint al final de cada iteración, esto nos lleva también a elegir la herramienta a utilizar, por mucho tiempo he utilizado Slack, pero si es de tu preferencia Microsoft Team u otro como Meet de Google no está mal, todo va depender de la complejidad del proyecto y lo que el equipo necesite compartir.

It is also very important to share the technical debt with your team. In software there are lot's of trade offs. The business teams need to understand clearly by doing functionality A it causes implications on business process B and C. It's a continuous process of communicating trade offs and technical debt with your team. A very important lesson I learn every day.

Con una buena estrategia las deudas técnicas son evitables. Muchas deudas técnicas son producto de una mala planificación, un inconveniente a solucionar rápido, o desconocimiento en la tecnología que se trabaja y sus buenas prácticas. Algunas deudas técnicas son asumibles dependiendo del impacto. Las deudas técnicas son oportunidades de aprender y especializarse.

Leyendo la opinión de LinkedIn considero que mas allá de un costo futuro de la repetición del trabajo adicional, yo enfocaría la Deuda Técnica basado en mi experiencia en la presión que ejerce el product manager o Scrum Master en cumplir plazos, en mucho de los casos ejercen esta presión pero no otorgan los recursos para priorizar ciertos sprint. La deuda técnica la he convertido en prioridad predictiva para luego pasar a lo preventivo y al mismo tiempo avanzar al estado correctivo. Falta de gerencia para identificar la deuda al igual que la autonomía en la toma de decisiones son las que he visto que generan la disminución de la calidad del software o la dificultad para agregar nuevas funcionalidades y quedar endeudados.

El aprendizaje continuo es un requisito para alcanzar un proyecto "excelente". En cada proyecto que se trabaja hay retos, y esos retos aportan lecciones. Es importante tener una fuente de información donde podamos documentar cada lección y que esa fuente sea común en caso de trabajar en equipo. Las apps de notas suelen ser muy útiles para este propósito. Cada lección aprendida nos permite ser mejores profesionales y aportar mejores soluciones en futuros proyectos.

La fomentación de aprendizaje continuo dentro del equipo es fundamental, esto se puede lograr estableciendo un horario de primera hora para inspirar, ¿de qué manera? compartiendo recursos, organizando sesiones de aprendizaje en conjunto y de manera individual. Ahora, puedes hacer todo lo anterior con tu equipo, pero debe combinarlo con técnicas del gestión del tiempo como Pomodoro, el método GTD (Getting Things Done) o la matriz de Eisenhower para optimizar tu flujo de trabajo. Antes de empezar una sesión de aprendizaje lo enfoco en áreas que como equipo se deben mejorar y la guinda al pastel de todo esto es reconociendo y recompensando el aprendizaje individual y conjunto de todo tu equipo por mejorar sus habilidades.

Understand the essential requirements versus the nice-to-have features. Focus on delivering what is necessary for the project's core functionality first. Be clear about what is within scope and what isn't. Scope creep can derail timelines and budgets, so ensure changes are properly evaluated and agreed upon. Keep stakeholders informed about progress, challenges, and decisions. Transparent communication builds trust and helps manage expectations. Be open to alternative solutions or technologies that can achieve the same goal more efficiently or within constraints.

As engineers, we tent to want to design the best algorithm in terms of speed, efficiency, elegance, etc. We want a O(1), but usually ‘good enough’ is just that. Focus on providing the most value to your customers in a timely manner. Complicated algorithms will unfold and get better over time. You will likely not have a perfect environment, nor will you have perfect set of requirements to build the ‘perfect’ algorithm for the feature in front of you. As you learn more, you can enhance the performance of your features and application.