Simon Hawke

I just watched Daniel Terhorst-North's talk at Kotlin conf, having sadly missed it in person (nothing to do with being hungover on day two of the conference). Daniel is the originator of Behaviour-Driven Development and gave a talk titled 'The best programmer I know'. A few interesting takeaways: - You don't need a degree to be an exceptional programmer - Everyone is a kinesthetic learner. Some people might lean towards seeing or hearing things, but everyone learns through doing, so just start and iterate as you go. Lean into uncertainty. - Invest in the product - not the code - Watch your users and find out what frustrates them - TDD pays off. Daniel's team at BP won a software quality award for their trading system. Unlike other teams, they found zero bugs during six weeks of UAT, thanks to TDD. BP paid them to stay on for the UAT period, which they spent in the pub. If that's not an incentive to learn TDD, what is?" - Explore different languages, tools and paradigms to give you more options and help you solve specific problems throughout your career - Help others - teaching is the best way to learn - Overall takeaway for me is that soft skills are clearly equally as important as hard skills in software engineering Worth a watch. https://lnkd.in/e8bau4X9

15

1 Comment

Throughout my career, by first or second hand experience I’ve seen two types of organisations: those that decided to take the leap and phase out architect roles and those that really cling to that. Based on what I’ve send and experienced, compared with accounts from my network and friends, the former type develops a culture where the architecture serves the code and the other way round, architects are replaced by a collective effort by senior engineers or principals. Every architectural decision are holistic choices based on the state of the stack, future evolutions (enablement), and what works best with the technology and programming languages in use. In the latter, architecture is often disconnected from the code, one doesn’t serve the other, and workarounds are a daily trade. There are degrees of disconnection, from a human and technological perspective where decision are taken from an ivory tower without much context of what happens or is needed in the ecosystems of services. One size fits all is often seen as a solution (a heavy process makes flexibility just a dream), any variations needs lengthy approval that discourage engineers from embracing a cycle of "modernisation by default”. Thoughts? #softwarearchitects #enginners #engineering #agile #modernisation

8

2 Comments

Now that the Microsoft AFO UK redundancy consultancy period has concluded, I can start to share details of where people are headed. Firstly there are people not actively job hunting for a software engineering role. - A decent number of people are taking the opportunity to go travelling for an extended period - A few folk are looking at changing career - e.g. starting a masters course, going into teaching, etc. - A bigger chunk have decided to take a couple of months off and begin job hunting in September. They'll have an easier time then, because they will have a stronger network.  - And finally let’s not forget the few that are still processing the shock; spare a thought for them. For the majority who are actively job-seeking though, early results are in. I'm thinking of it as like election night, there is excitement around those initial results, and we can start to form narratives, but they won't necessarily reflect the final picture. The job market dynamics will likely shift after the summer, interview process lengths vary by company size, and my guess is that those taking a relaxed approach may find themselves in different roles compared to the early movers. Nevertheless, here's the landscape so far: - Startups, with available funding, have moved the quickest  - Medium-sized companies are beginning to make offers. - Larger companies, with longer cycles, are interviewing, but not yet complete. Except Nvidia, who have hired a big bunch. Here are some specific destinations I've heard about within Engineering, with a very strong bias towards those "close" to me.  evroc (2) Oakdoor UNDO Squarepoint Capital Hymans Robertson Ciena Kraken NVIDIA (10+) Technical consultancy Microsoft (2) Update 5th Aug: NVIDIA (5 more, making 15 so far) Deliveroo (2) Cisco (1) Checkout.com (2) Cellular Origins (1) ---- Outside Engineering, I'm less in touch with where people are at, but I've noticed posts of people moving towards existing customers:  Award Consulting (2) weavix https://lnkd.in/egnSKVYd If you have more updates to add, please share in the comments!

171

12 Comments

It's interesting to see people so pressed about the Levels podcast. Of course there is a time to purchase tools and at certain scale it makes a lot of sense to optimise for speed and enterprise support over cost and customisability but there are far too many people entering the industry an only learning how a specific tool e.g. serverless Vercel hosting works over spinning up a VPS and learning the whole journey. This is leading to a trend where hard infrastructure and engineering problems are going unsolved except by the most senior engineers. Every couple of months there is some big post about how a serverless company has "ripped off" a person who had some success with their application. These companies are offering open source software to sell the platform on the backside which is exactly where they should be but consider if you are at the point where locking into and paying for that ecosystem is where you should be as an engineer or small business. It is still my belief that people new to the industry should focus on the fundamentals of how the underlying thing works even if using a tool like Next (I pick on Next a lot) because hand waving away all the complexity and hosting does not help in the long term. When I started out I would hear comments that Rails or Python were not "serious business" because they were not low level enough which I am sure is how I sound now but really consider when starting your business or indie hacking if you need the latest and greatest resume building tools or if you need to just build the thing in PHP and jQuery to get revenue. Ecosystems that do everything for you are great but if you try to work outside of their happy path you can have issues. Build your independence and learn how things work. You can watch the episode here: https://buff.ly/3YVp7t2 #softwareengineering #engineeringmanager #PHP #jQuery #NextJS #serverless #entrepreneurship #founder

4

I've started a long-deferred read through of Team Topologies, which speaks to some of the problems I was talking about recently around the ways "utopian" software development maxims can fail when they encounter the "topos" of an organisation's incentive and communicational structure. There's a thought-provoking section on John Sweller's 1988 breakdown of "cognitive load" into three categories ("intrinsic", "extraneous", "germane"). The thought it provoked in me was that I tend to disregard the burden of the "intrinsic" when looking for ways to reduce "extraneous" load (ways a codebase and development environment distracts me with boilerplate and arbitrary irregularity) and gain traction over "germane" problems (those particular to the domain we're working in). For me, leveraging advanced language features to eliminate boilerplate and build out a little embedded DSL that can express the problems I care about is an obvious win - I often refer to Paul Graham's maxim that you should first create the language you want to solve your problem in, then solve your problem in it. The problem with this is that it can shift extraneous load into intrinsic load, by requiring programmers who were comfortable with repetitive constructs using basic language features to familiarise themselves with much more advanced features (and the DSL's idiomatic usage of them) for the sake of eliminating what may seem to them to be quite harmless repetition. This suits me personally, because I like having an advanced model of a programming language in my head and using it creatively, but one isn't accounting properly for the distribution of cognitive load if one doesn't also take "intrinsic" load into account. I think of Golang as an intentionally low-intrinsic-load language - it has a small kernel of features, and is structured in a way that discourages higher-order abstraction over those features. If in doubt, write another for-loop! By comparison, I remember Scala adoption at TIM Group being an intensely painful experience for some of my colleagues, because it introduced so many constructs and type-level features that could be used in so many colourful ways. While I was elated by its endless opportunities for minimising extraneous load, they felt the burden of ballooning intrinsic load, and the frustration of being temporarily deskilled by the change of fundamental environment. The real art is to leverage small changes of technique and perspective, that will fit comfortably in everybody's head, to chisel away at extraneous load, until they feel that the nuts-and-bolts of their tasks are simpler than they have ever been.

12

I've seen a lot of contract roles posted over the last few months (increasingly Outside IR35 which is encouraging but not the point of this post) that list 'Azure' or 'AWS' as required skills .... which is fine to a point but both have over 300 individual services. Which aspect of these platforms are actually required for these roles is anyones guess but I'm probably not alone in thinking it would be useful to know. Azure: https://lnkd.in/erx4mWFZ AWS: https://lnkd.in/e_HnvuuJ #contracting #cloud #skills #clarification

7

16 Comments

SRE vs DevOps... I once got into a bit of a debate with a manager over the differences between SRE and a DevOps Engineers. His view was very much "they're one in the same" which I couldn't wrap my head around... It very much felt like a spotlight moment... Like I had Chris Tarrant sat opposite me (Sorry Jeremy). My stance and understanding from my time is kind of, what it says on the tin... Ensures uptimes, availability and robustness (i.e. reliability) of the Product and it's Infrastructure. His was very much no they need to be able to work across multiple domains and be able to code. As a consultant, Shock... I'll consult to partners what the market dictates, trends and help improve their processes but I've known a few DevOps and SRE interviews ask for coding tests and I often have to ask what the actual requirement is and if it dictates that? I get it if that's required and that skillset is necessary but then I would question calling that an SRE... I know it's always up for debate but keen to hear other peoples two cents or if I've got the wrong angle here... Alec Bruce-GardyneKim O'CallaghanBecca CoombesConnor AndrewsMaryann McGeeSteve BowenGeorge AldridgeAdam HolmesRichard Morrish

31

19 Comments

Here's a concise representation of your ML platform, courtesy of Laszlo Sragner.

3

A management conundrum. Last night I met up with an old colleague and the conversation turned to code written by super intelligent developers. Code which is extremely clever, but at the same time very difficult to understand and manage. Code that the average developer cannot maintain, and which even senior developers will take an inordinate amount of time to understand. My role as engineering manager is, amongst other things, to ensure the viability and maintainability of any software that is released into production. I am of the school of thought that software should be written as simply as possible, and that clever code is in fact not clever at all. I would like to discuss techniques on how one might manage said super coder, encouraging them to reduce the complexity of their code. The obvious initial steps are to discuss with the coder the impact of their clever code and attempt to get their buy in for the maintainability of the code, but where do you go if this coder considers simpler code as inefficient, boring, beneath them or any other reason? I would further float the potentially controversial opinion that the company might be better off letting the individual go if their codebase remains unmaintainable by 90% of developers.

9

5 Comments

I think I may be a little late to the party here but, blimey Charlie testcontainers.com is super useful for your integration tests. 1) I can see why Docker snapped up the developers, AtomicJar 2) testcontainers.com may make building integration tests just a little too easy

6

I was recently on the Reliability Enablers podcast with Ash Patel about my story, who I am and what I do. This is probably a good introduction for SRE's who may not have ever heard from me. We talk about how I became a performance engineer, the music I do for The Performance Tour, and my thoughts about NFR's and the importance of performance engineering. Thank you Ash for having me on and letting me rant. 😀 I'm not saying I am right about everything, but these are just my opinions. Hope you get something out of it. https://lnkd.in/g7anYRgH #performanceengineering #softwareengineering #performance #devops #SRE #sitereliability

8

2 Comments

I’m super excited to dig into this book, #FlowEngineering! It arrived today! “ as scale increases, waist and delay grow exponentially, and interdependence exacts a massive tax on the business” “We need to dismantle and decouple crippling dependencies and enable effective descaling to improve flow.” Here’s to improving flow!

31

10 Comments

Contractors or FTE's? (This post is not powered by chatgpt) I've been in tech recruitment for quite a few years now, and there's one debate that never seems to die: contractors or full-time engineers? It's a question I've seen countless hiring managers grapple with, especially as the tech landscape keeps shifting due to the economy, and other "creative" solutions now accessible. Look, I'll be upfront - I've worked with contractors a lot so generally have a bias towards. But even forcing a neutral view, I still can't help but think they're often the smarter choice for scaling teams. Here's why: - They're quick off the mark. Senior contractors are in and delivering before the ink's dry on their contract. With contractors I've seen projects kick off within days rather than weeks. - They're surprisingly wallet-friendly. No need to think about pensions, sick pay, holidays or kitting out an office. A client we work with in the US mentioned they saved 40% on overall costs by going the contractor route. - You can fish in a bigger pond. Top companies can work with the best engineers from across the globe. A team I staffed with had devs from 8 different countries, bringing a wealth of diverse experience to build a killer product. - They're flexible. Whether you need to scale up rapidly for a big project or downsize due to market changes, contractors offer agility that traditional employment can't match. - Efficiency. Exposure to various challenges across different companies often results in contractors having a broader skill set and more innovative problem-solving approaches. - What you see is what you get. Good contractors come with solid portfolios of projects and references. You can often try before you buy (contract to permanent hire if all works out), which takes a lot of the risk out of hiring permanently from the off, which is always a cashloss risk (paying notice periods, training etc. Don't get me wrong, I understand the appeal of building your own internal team. But in today's tech world, having a flexible workforce can be a real game-changer - your competitors are likely to be taking advantage by hiring contractors when you're not. I'm curious.. any contractor stories, good or bad? 😅 #TechHiring #ContractorsVsPermanent #SoftwareEngineering

44

6 Comments

Aside from automation skills that seem to sought after by most within QA + the ever evolving topic around AI, there's plenty of other ways of working/testing techniques that I hear mentioned consistently. - TDD, BDD, shift left, contract testing - to name a few.. Considering we're in a time that is so, so competitive - the last thing QA/QE people want, is to be working in a way that isn't sought after when they look for a change. If you were a junior starting out in your QA journey - what are the techniques/methodologies you'd focus on most? #QA #QualityEngineering

20

15 Comments

I thought it might be fun to start a bit of a LinkedIn Challenge - to meet 100 Developers in 100 Days I've been a C#/.NET software developer for over 2 decades. I've actually been coding with the .NET Framework since one of its beta releases Post a comment of what your primary coding language is and how long you've been mastering your craft #softwaredevelopers #coders #100devsin100days #linkedinchallenge

2

I love "translating tech" to non-technical colleagues at HelloBetter! If you become this person at work, you will make LOADS of contacts and understand your business better than ever. 🚀 Here's how it goes... You might remember I got 'Twitter famous' originally for my content and book for newbie programmers, which reflects the fact that I love to teach tech. 🥰💻 I've found that also in my working life, being able and willing to be the go-to person for basic tech questions has helped me a lot in making connections and expanding my understanding of a business. Why? 🧐 Because when you're the person ANYONE can come to, you meet colleagues from the entire company and you also get to know what they do. This not only helps on a personal level (deeper business understanding) but also helps professionally, since now if you're dealing with a system which touches their work, you have a point of contact to help you better understand it. I'm now about one month into my time at HelloBetter, and here are THREE examples of ways I've already done this: 1️⃣ I gave a presentation at our All-Hands about the basic evolution of engineering work from waterfall to agile, giving an overview of how creating software has changed. After this presentation, I stated directly that if anyone has any 'silly questions' about tech, they can come to me as I'm good at turning technical concepts into something anyone can understand. 2️⃣ When discussing testing with a Product Manager, they asked me to explain unit tests vs E2E testing vs smoke tests. I used the example of cooking a bolognese: "The unit tests are making sure that each step works, like that putting water in a pot on the stove will boil. The E2E tests are making sure that when you put those steps together, you actually end up with bolognese. The smoke tests are making sure that nothing else in the kitchen catches fire in the process." 3️⃣ I was discussing measures to take in response to an incident with one of our compliance team. She admitted that tech, to her, is like a "black box" in the organisation. I asked her if she wanted me to give her a run-down of the basics of what we do, and explained my team's composition and the differences between frontend, backend, mobile, and QA engineers, including a basic analogy for an API (the postal service). By doing these things, I not only get to do something I love (spreading a knowledge and understanding of tech), but I also get to know people I might otherwise not connect with so deeply. Better still: this is also good for your company! When different departments understand each other, working life becomes much easier 😊 #engineeringmanager #technewbies #programming Picture: Me geeking out at Perlan Museum on my recent Icelandic holiday, at their interactive exhibit on volcanic eruptions which occur beneath glaciers! Teaching and learning is fun 🤩

49

4 Comments

Very good insight. I have been working in agile development for almost 30 years, and would not want to go back. In fact, I almost changed career, and decided to stay solely because of Agile processes. It has become the norm, and newcomers just learn from what is in place: project management, development tools and practices. Agile is not dead, but I think the promotion of Agile as something new or different is. Agile is just how things get done these days. Even the worst functioning teams I have seen in the past 5 years is vastly more efficient than the average from 30 years ago, consistently delivering software with more value and fewer defects. Just like with any framework, not every team or enterprise has the same level proficiency in their process, or how to improve it. We should talk about continuous improvement and share techniques on how to evolve teams and environments, but talking about Agile to people who never lived through hard core Cascade development does not resonate with them. It is trying to solve a problem that no longer exists. 

7

3 Comments

developrec yourself session this morning was spent doing a bit of a comparison on Microservices and Monolithic architecture, a lot of our partners over the last few years have opted to break down Monoliths into Microservices but when is the right time to use Microservices? From an Application Architecture point of view Monolith seem to have a straightforward structure whereas Microservices are more complex. The benefit of zero-downtime deployment and CI/CD automation is attractive. I ran a poll on the benefits of Microservices last week and 86% of voters suggested that the ease of scalability/ability to scale independently was the biggest benefit. I would love to hear my networks opinion on the pros and cons of Microservices and when they should be utilised.

8

1 Comment