t-rav

Someone read uncle bob's clean architecture

right on, break the habit and focus will follow.

Pick a stack and go, your just wasting time and effort that is not focused on your value prop.

Strictly pair program the implementation to facilitate quick knowledge transfer of the tech stack.

In the early stages, it is about finding a minimal solution to fit your value prop so you can get people to engage with it and validate your assumptions. This requires more thinking than code. Pairing the implementation will not only allow tech knowledge transfer, it will facilitate those important idea discussions that need to happen too.

I agree with your boss, 40h work weeks are the right way to do things. It's up to you to find the discipline to focus during the day to make it truly productive, not just feel productive.

Finding a good workflow for what you do is the hard part. A good workflow will include timeboxed sessions between 50-90 minutes with regular breaks after each session to keep your mind fresh and focused. A lunch session away from the PC, take a walk. And a firm commitment to sticking to a true 8h work day, the real mental strain happens when one tries to exceed this daily limit.

And most important factor required to only working an 8h day, is to keep the messaging and social media use to outside work hours. These nasty distractions will destroy productivity and focus likely causing you to "work" and +8h day. All the chatting and social media can wait until you are home, life will carry on. Every distraction you give into will cause you to take 20 minutes to rebuild focus.

Why re-invent the wheel ! All you need to do is properly understand your tool of choice. I have used Neo4j extensively for years and have had great results. Our biggest hurdle was understanding how to model the data for a graph properly. Get this right and everything falls into place. The one thing you are going to need is a good visualisation tool, so be prepared to roll one of these too :)

Not sure this is what you are looking for, but I use Sonar to id code quality issues. Check it out @ http://www.sonarqube.org/

Use a budget burn down - That is set a fixed budget for a set of proposed changes. Where the changes are a mixed bag of critical and nice to have features.

Then promise to deliver some value. Where some value is the core features + some of the nice to haves. Promising or even trying to deliver a fully fixed set of features on a specific date almost never works in software development. Issues always evolved while doing the work. So plan on using a team lead who manages the project and sharp shoots issues with the dev team through paired programming. DO NOT PLAN on your lead having full development capacity. They will need to commit to other business activities to keep things moving.

Avoid over architecting the solution up front. Find the simplest value adding feature and work from there letting your design evolve. Back up your development efforts with TDD. And engage with the client / product owner often. This means frequent releases and meetings.

And at the end of the day delivering some value the client / product owner can leverage allows them to keep moving their objectives forward while the dev team evolves it understanding of the domain and challenges related to the project.

Finally, do not be afraid to say NO. If you feel it is unreasonable then say so. Avoid taking on work with insane deadlines no matter how important / critical it seems. Make a plan to work around the issue in a reasonable manor. Hopefully by continuously engaging with key stake holders issues are raised early and handled in a sustainable manor.

Success is all about consistency in setting and achieving goals. Take some time each evening to read, research and practice improving your skills in both tech and business. It may or may not yield the financial reward you are looking for, but it will yield greater personal satisfaction.

The only thing stopping you from achieving your goals is you. There is no magic formula, only hard work and perseverance.