Andres Rutnik

... what he said! ⬇️ There are many reasons why housing is vastly produced only by the private sector (read "Home Truths" by Carolyn Whitzman for more on that.) Spoiler alert: our housing crisis began when governments quit building homes. What we’re left with is a complex market propped up by prosperity, immigration, foreign investment, and cheap money—none of which are in abundant supply right now. So until governments or other players step back in to build, blaming “greedy” developers misses the bigger picture: private companies were never designed to solve social needs. Any yet: we're expected to.

4

Track nothing and you fly blind. Track everything and you risk nudging teams to optimize vanity numbers instead of real customer value. So our Engineering Management crew here at Vendasta headed to the whiteboard and asked: Which numbers would actually help us ship better software, right now? We set two ground rules: * Start small, refine fast. Perfect‑on‑paper metrics that never get implemented are useless. * Coach with data, never grade with it. Dev‑level stats exist to spark healthy habits, not to rank people. Here are just a few of the starter metrics we’re exploring: * Ramp‑up velocity ⏩ - How quickly do new engineers reach full productivity? * Roadmap vs. reactive work 🎯 - Are we spending capacity on strategic goals or fighting fires? * AI adoption & impact 🤖 - Who’s using the AI tools our org is standardizing on, and is it actually moving the productivity needle? * Hiring‑pipeline health 🏃‍♂️ - Time‑to‑fill, drop‑off rates, and how that aligns with the roadmap. This is just a taste. There are plenty more on the board, and we’ll trim or add as reality teaches us. To turn ideas into something useful Megan Cheesbrough (Engineering Manager), Lyle McRae (our new Senior Data Engineer), and I have started jamming on a proposal: a BigQuery + Dataform backbone with clear data standards so we can plug sources in once and let AI workflows keep everything fresh. Nothing’s built yet, but Lyle’s bringing proven patterns from past wins to shape a framework that fits our needs. Big unlock if we get it right. Our Principles: * Value first, vanity never. * Automate the grunt work; manage the AI. * Tell a story the whole org cares about. We’ve kicked off an MVP and will iterate. Your turn: What’s one metric that genuinely changed how your engineering org works? Drop a thought below. I’m keen to compare notes as we turn this whiteboard into reality.

21

4 Comments

Toolchain, build systems, SDKs, and anything similar are complicated beasts. Not every vendor gets them right. Midstall Software is committing to make using and building them better. By using open source software, this is very possible. And of course, by using Nix, this is fully reproducible and declarative. That makes scaling this between one developer's setup to an entire organization's worth possible.

7

1 Comment

Love this framing from Shopify. - stop debating AI adoption, start measuring outcomes. I’ve seen too many teams focus on getting people to use a tool instead of what it’s meant to solve. At Pebl, we use AI to make work easier — so our teams can focus on solving harder, more meaningful problems.

28

1 Comment

In a role like mine that bridges many gaps between development and production, I've always had trouble explaining what my actual job entails when someone asked, which generally led me to keeping it simple by saying "I create automation" and only going more in depth if they ask. That got me thinking, how would AI describe my job in a way that makes sense even to a 5 year old? Well... that led me down a fun AI chat rabbit hole that ended up with me becoming a super hero. I think I have a new favourite way of describing my role. Here's what ChatGPT came up with for my job description, including my very own movie poster. Movie trailer coming soon 😝 Name: The Cloud Crusader ☁️🦸‍♂️ Superpowers: - Summons invisible robot armies that can build whole cities (a.k.a. cloud systems) in seconds. - Speaks many secret robot languages — Bash, Ruby, JavaScript, Groovy — to give perfect instructions. - Uses legendary artifacts like Chef (the magic cookbook), Jenkins (the tireless foreman), Helm & Helmfile (the enchanted ship wheel and its map), Rundeck (the master schedule), and Terraform (the world-forging hammer that shapes the very ground the cloud stands on). - Can rebuild an entire kingdom exactly the same way, over and over, without a single brick out of place. - Keeps the cloud realm safe by making sure nothing breaks when heroes push shiny new buttons. Tagline: “By night he hunts bugs. By day he builds empires in the clouds.”

32

The Claude Code team uses Claude Code to build Claude Code. That's super meta. Last night at Volta, Rishabh A. hosted Halifax's first Claude Code meetup with support from Anthropic. We learned, demoed, and nerded out about claude code. I highly recommend checking the tool out. Rishabh kicked things off by showing how someone with technical background but limited coding experience could build entire slide decks and working apps. Then the Claude Code engineering team jumped on to talk about their product. They are fully dogfooding their product. They're using the exact tool they're building to build the tool.... Which means every friction point they feel, they can immediately fix. Great incentive alignment. Matt Cooper's demo was the highlight for me. He walked through his full-stack setup using Claude with skills and MCP. Here's the key insight on skills: you can create focused skill sets for different scenarios. One for frontend design, another for writing tests, another for library functions, etc. It's like giving Claude specialized knowledge for specific contexts, which reduces hallucinations and keeps the coding session on track. Matt also runs test-driven development to ensure new code doesn't break existing features. During the event, It was great to catch up with Abdul Samad, Omar Dahleh, and Ahmed Abdelmoaty. We talked vibe coding setups and LinkedIn posting. You know, the essentials. Halifax's AI community keeps showing up. Events like this prove it. If you haven't explored Claude Code yet, it's worth checking out!

72

7 Comments

For the past several years, SAS has been running a yearly Tandem Apprenticeship with 10 junior software developers at Code the Dream (CTD) who then are put in a hiring pipeline to continue their work with the company. We couldn't have asked for a better partner than SAS. This program allows SAS to truly test the CTD developers (similar to an internship), and ends with them getting 10 awesome developers at the end of it from diverse backgrounds, bringing diverse perspectives and a lot of talent to the team. If you think your company could benefit from this or have questions, please go to https://lnkd.in/e9QTZ8Pd or reach out! I'd love to chat :)

5

Incredibly proud of this team and the work highlighted here! Quantum Metric has been an absolute game-changer for our Digital teams, helping us deeply understand customer behavior, identify friction points, and build a better mobile app experience. But what I’m most excited about is how we've pushed the boundaries of this platform beyond traditional digital experience use cases, thanks to an incredible partnership with them and Google Cloud. Because Aritzia’s data platform and QM both live natively on BigQuery, we were able to seamlessly merge that data with Google Analytics 360 and the rest of our platform data. From there, we incorporated it into our existing server-side integrations via Hightouch to power channels like paid media. To top it off, for select downstream systems that require absolute speed—like personalization—we are using BigQuery Continuous Queries. This allows us to send critical data in near real-time, with only a few seconds from the time of the event until it lands in those platforms. Thank you to Sean De La Torre and the team at Quantum Metric for being such a forward-thinking partner, and to Nick Orlove and the BigQuery team at Google Cloud for the infrastructure to make it happen. Congrats to everyone involved in raising the bar! 🎉 #QuantumMetric #GoogleCloud #BigQuery #DigitalExperience #MarTech #RetailTech

104

4 Comments

Ambiguity kills software projects. Over-communicate, because chances are you are not communicating enough. Communication silos are more common than you think. If your project manager thinks of the software or goals differently than the developers, designers or clients do - you got a serious problem. Clarity helps exceed and meet expectations. Aka keeps people happy.

7

When Canadian metal band Spiritbox set out to document their first world tour, content chaos followed. Dropbox centralized their entire workflow, keeping every photo, video, and file organized, accessible, and ready to share. Read the full story: https://lnkd.in/gPGJZBa8

3

Software engineering isn’t just about shipping code—it’s about shaping teams that scale, systems that adapt, and cultures that endure. One of the biggest lessons I’ve learned in engineering leadership is that technical debt is easy to see, but organizational debt is silent — and far more expensive. We spend hours debating frameworks, performance tradeoffs, and system design. But how often do we apply that same rigor to the design of our team processes, feedback loops, or decision-making velocity? Here’s what I’ve seen work: • High-context environments beat high-control ones. Engineers thrive when they understand why, not just what. • Velocity emerges from clarity. The best teams don’t move fast because they cut corners—they move fast because they don’t waste cycles on ambiguity. • Great engineering managers are multipliers. Not the best engineer in the room, but the person who ensures every engineer is at their best. If you’re scaling a team or facing complexity creep, ask yourself: Are we debugging our systems as actively as we debug our culture? #SoftwareEngineering #Leadership #EngineeringManagement #TechCulture #TeamScaling #OrganizationalDesign ⸻ We are proud of the culture and organization we’ve built in Amazon Vancouver. Come meet us! Want to discuss about these in person? See you next Thursday, May 15th at the Amazon Payment Products event!

53

1 Comment

Being the hero is one of the fastest ways to break an engineering team Early in my career I thought being a strong engineer meant saving the day Prod is down I will fix it Deadline is tight I will stay up late. Docs are missing I will just figure it out. It felt good People noticed I got credit for being reliable But over time I realized I was doing real damage When you constantly step in as the hero a few things quietly happen First systemic problems stay invisible If you patch things silently leadership never sees the broken process that caused it Second you become the bottleneck If you are the only one who knows how to fix things the system only works when you are around Third you block growth Other engineers never really learn how to handle problems because you already did it for them I saw this pattern over and over on teams I worked with and it is one of the reasons we built Aude Most performance systems reward hero behavior Late nights Emergency fixes Big visible saves They rarely reward the work that prevents fires in the first place The boring improvements The documentation The design decisions that reduce risk Real leadership is not about saving the day It is about building systems that do not need saving Sometimes that even means letting things fail so the real issue becomes visible and can actually get fixed It is uncomfortable But it is how healthy teams scale Curious who else has had to unlearn the hero mindset

12

#1 Building Features for Companies (in their tech stack) until I get hired by one of them Company: Wealthsimple Feature: AI Chat for your portfolio This is a series where I build MVP demos for companies, with design docs on what an integrated implementation would look like. 🔐 The Big Takeaway My assumption is that the major barrier to this feature existing is data privacy. I had a cool conversation with Claude exploring potential solutions (doc in the repo in the comments) — most were data sanitization-based, but I think there's real room for enterprise AI to step up with FinTech/Banks here. Excited to see what that solution looks like in the near future. 🛠️ On "Vibecoding" with Claude Code There's definitely a sweet science to using Claude Code as an agentic partner to write code. Was AI slop written by the agent? Yes. Can it be drastically reduced with solid engineering principles in your design docs? Also yes. My current approach: leverage my Software Engineering knowledge to lay out concrete plans the agent can follow with oversight from me. This exercise proves to be a positive feedback loop that essentially lets me deepen my understanding on why certain engineering practices are important vs what is unnecessary. Curious to see how this evolves throughout the series. 🤖 The Setup Towards the end, I used a descriptive prompt to spin up a sub-agent that helps with this series — does preliminary research on a company's tech stack, explores current implementation challenges, etc. Also saw the buzz on Twitter when Remotion was integrated with Claude Code (had no idea what Remotion even was before this) and ended up using it for the demo. Got cool feature ideas for a company? Drop them in the comments 👇 Mandatory CTO tag (to push myself out of my comfort zone and put myself out there pls don't hate me for tagging you): Diederik van Liere

34

2 Comments

Stop talking about AI for a moment people. A new economic order is being built from the ground up. Jobs (just like the [GTM] Collective said would happpen) are being shed by multi trillion dollar organiations... and everyone around them. Hiring / Job acquisition is looking a lil' different. Welcome to the first innings to post-labor economics, categorically being defined by leaders like #DavidShapiro. DTR (Developer Technology Research) is at the driver seat of post-labor economic digital infrastructure, including building application platforms for developers to contribute their skills, to collective OSS projects - ultimately enabling 'The Forge of Open Innovation' that I keep pressing on about. The operating model for developers to get a CS degree -> Internship -> Experience -> Stable Career is... obsolete. Post Labor economics needs a new infrastructure stack to shepherd in this 'new era'... and the collective we've built will ensure the system thrives in the chaos of disruption. Just don't be surprised when we look back in three years and point to these seminal moments when you had the opportunity to positvely impact the system. What if this all ends up being... True? What if data no longer had gravity? If AI alignment was built into the silicon itself? If 1B compute nodes became accessible across sovereign lattices?” What if we could represent 20GB files as 512 bit hashes...? What if I told you the [GTM] Collective & UOR Foundation has already solved, or is actively solving these problems? No I did not use AI to write / draft this. It's too important. Do not bet against the collective.

10

7 Comments

One of the more underrated things a Rails developer can do for their career is show up to an in-person event. The obvious value is learning. But the less obvious value is compounding. The conversations in the corridor between talks, the developer you get chatting to before a keynote, the hiring manager who notices your question during a panel — none of that happens on a Slack channel or in a GitHub thread. Sorcer were proud sponsors of Brighton Ruby last year, where we saw exactly this in action. The room was full of developers at every level, from juniors finding their feet to principals who’ve been writing Ruby since the early days. The quality of conversation across that range is something no conference livestream replicates. If you’re considering whether a developer event is worth the time away from the day job, the research backs what most experienced developers already know: career-defining roles are more likely to come from a hallway conversation than a job board. Brighton Ruby returns this year. If you haven’t been, put it on your list. #BrightonRuby #RubyOnRails #RailsDevelopers #DeveloperCommunity #TechEvents #RubyConference #Sorcer

6

After a chat with a great contractor yesterday, we got on the topic of T-Shaped profiles in contractors. In today’s fast-paced tech landscape, the most impactful contractors aren’t just specialists, they’re T-shaped. A T-shaped JavaScript contractor combines deep expertise in core JS technologies (React, Node.js, TypeScript, etc.) with broad capabilities across adjacent areas like DevOps, UX, testing, and architecture. They don’t just write code, they solve problems holistically. 🔍 Why T-shaped contractors stand out: Adaptability: They can jump into unfamiliar territory and add value quickly. Collaboration: They speak the language of designers, product managers, and backend engineers. Efficiency: They reduce bottlenecks by bridging gaps across the stack. Strategic thinking: They understand how their work fits into the bigger picture. Whether you're scaling a product, modernising legacy systems, or building something new T-shaped contractors bring the agility and insight to move fast without breaking things. 💡 Looking to hire? Prioritise depth and breadth. The best contractors don’t just code, they connect dots. #JavaScript #Contractors #Hiring #TechTalent #TShapedSkills #Frontend #FullStack #React #NodeJS #TypeScript #AgileTeams

31

1 Comment

You can’t do #PLG if every event needs engineering. That’s not Product-Led Growth. That’s backlog-led growth. 🧱 I was in #Vancouver last week and grabbed coffee with my friend Thera. We ended up talking about #PLG, and why so many teams say they’re doing it… but never really get it off the ground. Here’s the uncomfortable truth most teams avoid 👇 PLG doesn’t fail because of bad strategy. It fails because no one can get engineering time to track anything. Marketing wants activation data. Product wants feature adoption. Growth wants funnels. Engineering is busy shipping the product. So everything stalls. If your PLG tracking requires engineers to: 🔧 Add events for every click 🚢 Ship code for every experiment 🧠 Maintain analytics logic You’re not running PLG. You’re running a queue. I’m a technologist hell-bent on helping the business elegantly solve tech-adjacent problems. So here’s What actually works 👇 Ask engineering for one thing only: 🧩 Load a tag manager after sign-in 🔑 Pass basic user context (user_id, plan, org) That’s it. Everything else happens outside product code: 🛠️ #GTM for event injection 📊 PostHog for analytics and experiments No dev tickets. No redeployments. No waiting three sprints for a funnel answer. What this unlocks 🚀 🧭 Product defines events visually 📈 Marketing tracks activation without code ⚡ Growth iterates in days, not months PLG is about learning velocity. If every insight needs engineering time: 🐌 Experiments slow down 💀 Hypotheses die 🤐 Teams stop asking good questions Remove the dependency. Increase learning speed. Let engineers build product and marketers grow the business.

16

3 Comments

“Is TikTok safe for a 12-year-old to use?” If you’ve ever asked a question like this, you’re not alone. At KidsNanny - Parenting Tech, we’ve built a new tool to help parents make smarter, safer decisions: 👉 AI-Based App Reviewer – now part of our AI Safety Companion. This smart system uses multi-agent AI to analyze: App store descriptions Permissions requested Recommended age ratings User reviews And yes—your child’s age is considered in every evaluation 📱 Whether it’s a viral app or something new, get clear, instant insights on whether it’s safe for your child. 🧪 Currently in beta—we’d love your feedback! Try it now: https://lnkd.in/d_4rhX4W #ChildSafety #DigitalParenting #KidsNanny #AppReview #AIForGood #OnlineSafety #ParentalControl #BetaFeedback

8

3 Comments