Rhett Blanch
Greater Sydney Area
537 followers
500+ connections
View mutual connections with Rhett
Rhett can introduce you to 10+ people at Flynn Group
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
View mutual connections with Rhett
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
Websites
- Personal Website
-
http://www.wilgatree.com
- Portfolio
-
https://tallymarks.io
- Company Website
-
https://www.pizzahut.com.au/
About
Welcome back
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
New to LinkedIn? Join now
Activity
537 followers
-
Rhett Blanch shared thisData governance often gets seen as something that slows things down - but in reality, it can lay the foundations that help things move faster and more reliably. Reporting, insights, and AI built on inconsistent or undocumented data can quickly run into problems. Without clarity and alignment, it’s easy to end up spending more time fixing issues than gaining value. When there’s no data catalog or clear ownership, a lot of energy ends up going into resolving definitions instead of delivering insights. Strong governance doesn’t have to be a blocker - it can actually be a key enabler when done well.The vital importance of data governance in the age of AIThe vital importance of data governance in the age of AI
-
Rhett Blanch reposted thisRhett Blanch reposted thisI’ve learned this the hard way: A great data culture doesn’t start with tools or models. It starts with listening. When data teams listen, really listen, to the business, they hear: ✅ The real problems behind the requests ✅ The context behind the numbers ✅ The goals and challenges that data alone won’t show When business teams listen back, they learn: ✅ What’s possible (and what’s not) with the data ✅ The risks and trade-offs in decisions ✅ How data can inform, not just report True alignment happens when both sides listen with empathy, and solve problems together, not in silos. In my latest article, I explore how business and data teams can row together, building a data culture where collaboration, empathy, and real impact take center stage. ✨ It’s not about perfection or overnight change. It’s about small wins, continuous learning, and working together to create real value. #DataCulture #DataDriven #Collaboration #Empathy #TechLeadership #CitizenDevelopment #Analytics #MindsetMattersThe Magic Happens When Business and Data Teams Row TogetherThe Magic Happens When Business and Data Teams Row TogetherChetna Chaudhari
-
Rhett Blanch reposted thisRhett Blanch reposted thisEnterprise architecture (EA) practices, as complex organizational practices embracing multiple people, processes and documents, can hardly be viewed as all-or-nothing propositions and can rarely be established in organizations at once. Instead, different elements of an EA practice tend to be introduced and institutionalized in organizations gradually, by progressing from more basic to more advanced activities. But which elements of an EA practice can be regarded as basic to be adopted first and which as advanced to be attempted last? All the mainstream EA maturity models pretending to answer this question are mindlessly copied from the classic Capability Maturity Model (CMM) by mere rewording, do not reflect any genuine specifics of an EA practice in terms of its elements (or reflect them improperly) and, unsurprisingly, fail to elucidate the sequence of their introduction. To address this gap, I have just released yet another one-page deliverable titled “Enterprise Architecture Maturity on a Page” that explains the most typical (not to be confused with the “right”, “necessary” or “only possible”) maturity journey of EA practices observed across the industry: https://lnkd.in/eZVhgzZh. The proposed EA maturity model is aligned and connected with the three previous on-a-page products depicting an EA practice, EA function and EA artifacts respectively (all available for free here: https://eaonapage.com). The origin, logic and maturity stages of the model are described here: https://lnkd.in/eJetGJap (or here: https://lnkd.in/ezppWFVd). Enterprise Architecture Maturity on a Page is an evidence-based model of an EA maturity journey on a single page. It provides an aggregated one-page view of the maturity of an EA practice with its different stages and their essential properties, including their constituting processes, existing architecture positions, established governance arrangements, mastered EA artifacts, associated challenges and realized benefits. Enterprise Architecture Maturity on a Page is freely available to download here: https://eaonapage.com #EnterpriseArchitecture #EntArch #EAonaPage
-
Rhett Blanch shared thisIn his article “The Data Disconnect: Why Digital Transformation Needs a Data Strategy,” Jan Meskens highlights a critical oversight in many digital transformation initiatives: the neglect of robust data management. He observes that organisations often invest heavily in creating user-friendly digital interfaces—the “flashy front-end”—while underlying data management practices remain weak or underdeveloped. This imbalance can lead to significant challenges, including poor data quality, integration issues, and missed opportunities for leveraging data-driven insights. Good read.The Data Disconnect: Why Digital Transformation Needs a Data StrategyThe Data Disconnect: Why Digital Transformation Needs a Data Strategy
-
Rhett Blanch reposted thisRhett Blanch reposted thisCongratulations to the Isolated Children’s Parents’ Association (ICPA) on the launch of Rural and Remote Education Day (RREd Day) today, 9th October 2024; a dedicated awareness day to shine a spotlight on the unique challenges faced by students and families in rural and remote areas of Australia. “Rural and remote students face extraordinary challenges every day, from connectivity issues to limited access to educational resources. RREd Day is about bringing these issues to the forefront of national conversation, ensuring that rural and remote education is not just an afterthought but a priority. By wearing red, sharing stories, and joining forces, we can each work to drive the positive change that our rural, remote and isolated children so desperately need.” ICPA Federal President Louise Martin (QLD) Burgmann staff and volunteers wore it red today in support of the ICPA and to highlight the importance of access to education for students from rural, regional and remote Australia. Burgmann has a strong community of alumni and current residents from across Australia's heartland who are treasured members of the Burgmann family. For our 2024 Annual Appeal, our community is working to establish The Harriet Nixon Scholarship – an endowed scholarship for a rural, regional or remote students – in memory and honour of alumna Harriet Nixon (2015-2016). Harriet Nixon grew up on small rural cattle fattening property with her parents Ralph and Fiona Nixon and younger sister Alicia outside Moss Vale in the Southern Highlands of NSW. Harriet was a remarkable young woman who embodied the values and spirit of the Burgmann community, and whose legacy and memory continues to inspire. Read more here: https://loom.ly/i4vdE4I Give today: https://loom.ly/ZBSv3P8 Pictured: Fiona Nixon (Burgmann Volunteer, BCAA Member and mother of alumna Harriet Nixon) and Taryn Stasakova (Deputy Principal); and Fiona with Amelia Zaraftis (Director of Advancement). #WearRREd #RemoteAustraliansMatter #RuralRegionalRemote #TheHarrietNixonScholarship #BurgmannCollege #OutbackMagazine
-
Rhett Blanch shared thisOpenAI's custom GPT allows users to create and share personalised GPTs with specific instructions, and API integrations. This article provides a step-by-step guide to turning existing APIs into custom GPTs. It will be interesting to see how existing APIs can be leveraged to support new capabilities.
-
Rhett Blanch shared thisData governance goes beyond mere compliance and security. Proactive data governance is key to empowering a data-driven organisation. By establishing data catalogues, clear ownership and transparent policies for data access and use, governance programs not only meet regulatory requirements but also pave the way for effective data analysis to drive business outcomes.The secret to successful citizen data science programs: Good governanceThe secret to successful citizen data science programs: Good governance
-
Rhett Blanch shared thisRhett Blanch shared thisWe are delighted to announce that Allegro Funds has entered into an agreement to exit its investment in Pizza Hut Australia through a sale to Flynn Restaurant Group Flynn Restaurant Group is the largest franchise group in the USA, operating over 2,350 restaurants including 926 stores of the world’s largest Pizza Hut franchisee, Hut American, as well as hundreds of stores for popular QSR brands Applebee’s, Taco Bell, Panera, Arby’s and Wendy’s. Since acquiring the master franchisee from Yum! Brands in 2016, Allegro partnered with Pizza Hut’s CEO Phil Reed and its dedicated management team to execute a turnaround, successfully transforming the business into a profitable and growing network through an unwavering focus on operational excellence and customer satisfaction. We are excited for our future growth and partnering with the team at Flynn Restaurant Group in delivering world-class Pizza Hut experiences for our customers across Australia! Together, we will get back to number 1. https://lnkd.in/g72rtrKNPizza Hut Australia has been sold to Flynn Restaurant GroupPizza Hut Australia has been sold to Flynn Restaurant Group
-
Rhett Blanch shared thisRhett Blanch shared thisBuilding knowledge is often expensive and it's fairly common to see knowledge walking out the door and leaving a void behind, especially in large organisations. "International Data Corporation estimates that Fortune 500 companies lose $31 billion from not sharing knowledge within their organization every year." If you are working in an organisation where teams are unaware of what others are doing, you might be rediscovering knowledge over and over again. Here's my brief take on the culture and tooling aspect of this massive topic. #business #knowledgesharing #kms https://lnkd.in/gxP2C_DTEffective knowledge management to reduce the cost of learningEffective knowledge management to reduce the cost of learning
-
Rhett Blanch liked thisSharing this because I’m genuinely happy it’s finally live. But also because creating the content felt like a milestone in my journey with AI tools. I’ve written before about using AI as a creative partner. I almost always give it something to work with first. A draft, an idea, a point of view, or if I’m writing thought leadership for someone else, a transcript of a conversation. I never let it create anything from nothing. This was different. I had to dig deep to create this content. It needed heart. It needed to feel grounded in our history. It needed to sound like us. And while it was all new, it also had to be built from everything that already existed. That made it hard. Some days it loomed over me like a giant monster asking, “So... what’s happening here?” AI could help with idea generation, but the constant loop of idea, check, refine needed more energy than I had at some points. What it was best at, though, was helping me sense-check the work against the vision for the website and how we wanted to describe our services. I initially spent a fair bit of time saying things like: “what do you mean that's fine, it doesn’t have enough heart” and “wrong, that’s definately not in the same tone as that bit.” Eventually, I built a custom GPT and used it to check the content against the instructions as a final pass. That process has now inspired a series of custom GPTs to help define DiUS’ brand voice and scale that thinking across the team, from social content to thought leadership. So, ChatGPT. Couldn’t have done it without you.Rhett Blanch liked thisThere’s a new DiUS website. On the surface, it’s a much better, clearer way for us to show up on the internet. But getting there was harder than we expected. The breadth of what we do is… a lot. It has to be. The problems our clients bring us don’t come in neat shapes. They cut across design, product, data and engineering, with AI now more visible across all of it. Some are modernisation. Some are greenfields. Sometimes we’re building end to end, sometimes we’re brought in for something very specific. No two are the same, but the way we approach them is. Thoughtful problem solving, and bringing the right combination of skills and expertise to each one. Trying to represent that clearly, without oversimplifying it, was genuinely difficult. At the same time, we had to make sure we reflected the impact AI is having on how we design, build and deliver software. Not as something new for us, but as something that’s now shaping the work more visibly. That shift is real for the teams we work with. They’re being asked to move faster, absorb new ways of working, and still deliver systems that hold up. In that kind of environment, experience matters. Knowing how software actually gets built, where complexity hides, and what stands up in production becomes even more important. Now, the website has caught up to where we already are. If you know us, you’ll recognise it. If you don’t, it’s a good place to start. www.dius.com.au
-
Rhett Blanch liked thisRhett Blanch liked thisProfessor David Lindenmayer has spent more than four decades tracking how Australia’s ecosystems change over time. A conservation scientist and biodiversity expert at the ANU Fenner School of Environment and Society, his work focuses on how species and landscapes respond to logging, fire and land management. He leads multiple large-scale, long-term research programs across south-eastern Australia, some spanning more than 40 years. His research connects field-based observation with environmental management, helping shape how biodiversity is conserved across forests, farmland and protected areas. Alongside this, David has built one of the most extensive bodies of work in his field, while supporting the next generation of scientists and contributing to policy and practice. Read more about David: quicklink.anu.edu.au/lz97 ANU College of Systems & Society
-
Rhett Blanch liked thisRhett Blanch liked thisLast week’s NSW & ACT Awards Gala was a great reminder of what makes our network truly special — our franchisee partners and the incredible teams behind each store. What stood out most on the night wasn’t just the results, but the ownership, pride, and teamwork our franchisees continue to demonstrate across the market. It’s this collective mindset that drives sustainable growth and keeps us moving forward. A sincere thank you to all our franchisee partners across NSW & ACT — your commitment, resilience, and focus on delivering great customer experiences continue to raise the bar for the entire network. Congratulations to our 2025 award winners: • Premier Franchise Partner – Jack Morrison & Pranav Sharma (Emerald Hills & East Maitlands) • Care Genuinely for People – Ashish Sambher (Seven Hills, Mount Druitt & Baulkham Hills) • Play Like a Champion – Rafi Islam (Goulburn & Mittagong) • Win as One – Chetan Kalra & Harshvardhan Mehta Mehta • Build a Kick-Ass Team – Umang Patel (Tamworth, Armidale, Singleton, Grafton) • Take Pride in Your Place – Talvinder Singh (Collaroy) • Create Great Guest Experiences – Satnam Ahuja (Windsor) • Deliver Sustained Results – Mehul Patel (Dubbo) • Own It – Adam Leishman (Narellan) • World-Class Ops – Dhaval & Mehul Patel(Lithgow) Thank you to Michael Lofhjelm and Rhett Blanch for joining us from RSC, and to Sarah Flanagan and the NSF team for being part of the evening and celebrating success with us. And finally, a huge shoutout to our incredible Area Operations Managers — Keval Darji, Seb Bivona, Alex Woolfenden, Mohab Maggar, Pheng Toeng and Darren Macclure — for the effort, planning, and energy you brought to delivering such a fantastic event. Proud of the momentum we’re building across Pizza Hut Australia — and even more excited about what’s ahead.
-
Rhett Blanch liked thisRhett Blanch liked thisGrateful to have been on stage at Elastic{ON} Sydney last week. For too long, 'security scores' and 'ticking the box' have passed as evidence of security maturity. Choosing to measure success by reducing material risk instead is a harder, more deliberate leadership call, and the honest conversations at ElasticON showed that the other security leaders already know this gap exists. Now lets close it. #ElasticON #Cybersecurity #ctrlcyber #SecurityLeadership #ROC #RiskOperations
-
Rhett Blanch liked thisRhett Blanch liked thisIn 1962, Thomas Kuhn explained why the most experienced people in a field are often the last to accept revolutionary change. It wasn't because they were wrong about everything. It was because they were right about so much that they couldn't see what had shifted. I've been watching the discourse around AI-assisted software development for three years now. The intensity of the resistance from experienced, accomplished engineers — the manifestos, the adversarial benchmarks, the philosophical objections — has been striking. Not because the critics are foolish. Many are people I genuinely respect. But because the pattern is so recognisable. It's Kuhn's pattern. The same defence mechanisms, the same incommensurable worldviews, the same emotional temperature that accompanies paradigm shifts. And yet the technology work. The code ships. The adoption accelerates. I wrote a long essay exploring how Kuhn's framework maps onto what's happening in software engineering right now — from the anomalies that don't fit the existing paradigm, to the defence mechanisms, to the emotional core beneath all of it. It's written with respect for the old paradigm and the people who built remarkable things within it. But it's also honest about what the evidence shows. There has been a revolution in Software Engineering. https://lnkd.in/gR5HPuJr
-
Rhett Blanch liked thisRhett Blanch liked thisA lot has been happening in the development world. Job layoffs, AI model advancements, agentic models, OpenClaw, and every day, a new improvement hits the news cycle. I have been watching and paying close attention to the trends, what's happening in the job market, AI agents challenging established SaaS businesses and how AI models are evolving at a rapid pace. I wanted to test the hype rather than just consume it. So 6 months ago, I started using AI as an assistant on a daily basis, helping me autocomplete lines of code as I wrote them. The output was rudimentary at best. My trust was low. But 3 months ago, Claude Opus 4.5 dropped, and for the first time, I saw a glimpse of critical thinking emerging out of the model. It pushed back, it challenged with questions, and it reframed problems. It went as far as testing a code change against how the existing code would behave. Something has changed! That was the moment I saw my role start to shift, from writing code to directing the agent that writes it. Developers are not disappearing. Far from it. The job is evolving into something new. I wrote about what I found. Link in the comments. #AI #SoftwareDevelopment #AgenticAI #Developers
-
Rhett Blanch liked thisRhett Blanch liked thisThis week, I was interviewing a candidate and mentioned it was fine to use any tool to perform the coding part. The candidate said he had disabled all AI tooling thinking it wasn't allowed to use. My understanding is that AI tools are available to every engineer today, and I would love to see how candidates use them to solve problems. That is not cheating; it is being productive. Am I right to think that?
-
Rhett Blanch liked thisRhett Blanch liked thisThere's been a lot of focus on natural hazards over these last weeks - for good reason as we look upon the devastation from fires in the south of the continent and flooding in the north - and unfortunately that has meant "natural disaster" has just about hit saturation point. So I thought I'd pen this article, based on an essay I wrote last year, on why language matters and why it's time to retire the term "natural disaster" for good. P.s. Evolution is good. I go back and reread reports, correspondence and comments that I've written as recently as early as 2025 where I've used the term, and while I cringe now, I also accept that this is an unavoidable outcome of learning and growth.Why It’s Time to Retire the Term “Natural Disaster”Why It’s Time to Retire the Term “Natural Disaster”Will Barton GAICD
Experience & Education
-
Flynn Group
******* ******** ************
-
********
**** ** ******** ************
-
*********
****** ****************** *********
-
*** ********** ******** **********
******** ** ******* ********** ********* undefined undefined
-
*** ********** ******** **********
******** ** ***** ********** ******* undefined
View Rhett’s full experience
See their title, tenure and more.
Welcome back
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
New to LinkedIn? Join now
or
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
Licenses & Certifications
Volunteer Experience
-
Musician
Lane Cove Community Bands
- Present 6 years 8 months
-
Musician
Lane Cove Community Bands
- 4 years 6 months
Courses
-
Leading SAFe 4.0
-
-
Product Management Excellence
-
-
Secure Coding
-
Recommendations received
4 people have recommended Rhett
Join now to viewView Rhett’s full profile
-
See who you know in common
-
Get introduced
-
Contact Rhett directly
Explore more posts
-
Joel Xiong
BudgetWise (budgetwise.au) • 1K followers
Why "Headless" isn't enough: Security & SEO at the Edge Building a headless platform with Drupal and Next.js is a great start, but how you handle the "Edge" is what determines its production-grade success. For my work on budgetwise.au, I focused on a robust Nginx and Let’s Encrypt layer to ensure security and SEO aren't just afterthoughts. Here is the strategy: Automated TLS Termination: Utilizing Let’s Encrypt for automated SSL certificate management ensures the platform is always secure and trusted by search engines. Nginx as a Security Shield: Acting as a reverse proxy, Nginx allows for custom security headers and protects the underlying Node.js and Drupal services from direct exposure. Canonicalization for SEO Authority: I implemented strict routing and canonicalization logic at the proxy level to prevent duplicate content issues, a vital step for maintaining SEO rankings. Decoupled Metadata: By separating the frontend (Next.js) from the CMS (Drupal), we can deliver lightning-fast, SEO-optimized metadata and Open Graph images without the overhead of a traditional CMS theme layer. Security and SEO are two sides of the same coin: Trust. A secure site is a searchable site.
-
Niranjan Limbachiya
TryGrounded AI • 37K followers
I was speaking with a CTO of a fast-growing product company. Their pipelines were fast. Their automation was strong. Their team was sharp. But still… releases were getting delayed. Not because of code. Not because of bugs. Because of test data. 1) Inconsistent datasets 2) Missing edge cases 3) Privacy and compliance concerns 4) Time wasted generating and managing data And that’s when it hit me again… Test Data Management is one of the most underestimated bottlenecks in modern testing pipelines. We talk about AI. We talk about automation. We talk about performance. But without the right data, even the best testing strategy falls apart. At KiwiQA Services , we are seeing this pattern across enterprises — especially those building data-driven and AI-powered products. The complexity is real. And the impact is massive. If you are building products where data is at the core, you cannot afford to ignore this layer. Curious to learn more? Read here: https://lnkd.in/g5-cRbMs And if you are going through this challenge or building something tough where quality truly matters Let’s talk. We would love to help you build confidence into your testing strategy. #Testing #AI #QualityEngineering #TestData #KiwiQA #SoftwareTesting #Automation #DataDriven
8
-
Muhammad Hassan
CodexPro • 4K followers
The next 18 months will define NDIS sustainability. Without a digital-first overhaul, the scheme simply won't scale. The current infrastructure is buckling under administrative weight. We are seeing a system trapped in the past: → Manual processing 📝 that delays critical support for participants. → Fragmented data silos 📉 making fraud detection nearly impossible. → Lack of real-time tracking ⏱️ for budget utilization and outcomes. → High compliance costs 💸 that drain resources away from frontline care. The problem isn't a lack of funding. It’s a lack of modern architecture. 🏗️ We cannot manage a multi-billion dollar social scheme with 20th-century tools. True sustainability requires: → Automated claims 🤖 to reduce human error and overhead. → Interoperable systems 🔗 for seamless provider-participant communication. → AI-driven insights 🧠 to identify high-risk spending patterns early. Technology isn't just an "add-on." It is the only way to protect the scheme's future. 🛡️ Is your organization ready for the digital shift? #NdisSustainability #DigitalTransformation #DisabilitySector
1
-
Roland Lopez
ODDS-TEAM • 3K followers
SQLite is not the default in rails for nothing Remember when I said under $200/mo, Kamal on a single box is the move for solo founders? I understated it. SQLite + Kamal + Rails 8 is a bomb improvement in DevOps. ECS, Fargate, RDS, ALB, NAT Gateway, VPC with multi-AZ subnets, bastion host, Secrets Manager, VPC Endpoints. 16 Terraform files. ~$80-120/mo minimum. Here's the same app with SQLite + Kamal: - One EC2 instance. - Kamal Proxy handles SSL and reverse proxy. - SQLite sits on disk Zero network latency, zero connection pooling headaches, zero database subnet configuration. $18-22/mo all-in. What I dropped: - RDS -> SQLite on disk - ALB -> Kamal Proxy (built-in) - ECS/Fargate -> Kamal deploys Docker directly - NAT Gateway -> gone ($32/mo saved just there) - Bastion host -> kamal console - Secrets Manager -> .kamal/secrets - VPC Endpoints -> no private networking needed - ECR -> GHCR (free) - Terraform -> config/deploy.yml (one file) Rails 8 ships with Solid Queue, Solid Cache, and Solid Cable — all backed by SQLite. Your background jobs, your cache, your WebSockets — all on-box, all file I/O. No Redis. No Postgres. No managed services. Your backup strategy is EBS snapshots + Litestream to S3 if you want continuous replication. Done. Now — I'm not telling anyone to migrate off Postgres. If you're running Postgres in production and it's working, there's no point moving. Postgres is battle-tested and incredible. But for new projects? I'm starting with SQLite.
9
-
Artavazd Balaian
Agoda • 1K followers
Very well written post on how bun tackles performance problems of existing JS package managers by Lydia Hallie. “bun install” is super fast because it treats package installation like a systems programming problem, not just JS scripting. Many of the slownesses in npm / pnpm / yarn stem from assumptions and abstractions that were valid years ago (slow disks, weak CPUs, limited memory), but are now bottlenecks. Bun’s goal: cut down overhead at every layer (syscalls, JSON parse, memory allocation, cache line friendly data structures, file copying, etc.). https://lnkd.in/g2wKcRBu #bun #js #performance #zig
7
-
Refat Rakhmatulin
UiPath • 852 followers
As software delivery gets faster, testing becomes more important, not less. Industry research has put testing / QA at roughly 23% of IT budget, and likely higher in reality once you account for the testing effort embedded across engineering and delivery teams. Meanwhile, MBIE says 67% of larger New Zealand businesses now use some form of AI. Faster change is here; the question is whether testing is keeping up.* That’s why this Auckland breakfast next week is timely. The discussion will focus on agentic testing and what it can mean in practice for enterprise teams trying to reduce manual effort, improve coverage, and move faster with more confidence. The roadshow highlights AI-generated and self-healing test cases, 50-70% faster regression cycles, coverage across SAP, Oracle, Salesforce, web and legacy systems, live demos, and peer discussion with QA and engineering leaders. UiPath’s Global Product Lead for Testing will also be joining in person to share practical insights from the field. The Auckland event is schedulled for Tuesday, 17 March 2026 at Park Hyatt Auckland. If this sits in your remit, it should be worth a look. Event registration: https://lnkd.in/e-5YfxuA *Capgemini - World-Quality-Report-2019-20 #SoftwareTesting #TestAutomation #AgenticTesting
10
2 Comments -
Ilya S
Netflix • 883 followers
Prompting fatigue and buying the dip I have a simple support task sitting in my queue today: export data from prod and import it into test to enable UAT for a stakeholder next week. This is one of those use cases where tool coverage is sparse. We have a tool for exporting, but not one for importing. I need to glue together a few API endpoints to make this work. No big deal. Especially when you have Claude, right? Well... wrong 😀 I am at the point where my excitement about being able to quickly vibe-code a solution to any problem is fading. Claude is great, but it needs to be told what to do. For that I need to: - allocate time - write up the requirements file - let Claude execute - review the result - test This is still work, and it is arguable whether Claude lifts the majority of the effort here. People, including me, just want solutions. Even if building a solution is easy, we still don't want to spend time filling every pothole. If a specific pothole is large and frequent enough, filling it becomes a business opportunity for someone else. With its risks, rewards, and glory. This is what SaaS is for. But SaaS stonks are down lately, largely due to the sentiment that everyone can build their own CRM, ERP, ATS, and so on. When I look at this, and then I look at that task in my queue, I smile and I buy the dip 🙂 Not investment advice. Consult a professional, or at least your mom, before buying any stonks.
1
-
Benedict Udemeh
TribeAnalytics • 166 followers
It's been a busy day but won't stop me from completing the how-to series for this week. 🧪 How-To Friday: Test Your Auth Like a Hacker, Without Causing a Panic. Nothing breaks trust faster than broken login flows. You push a new SSO config. You roll out token refresh logic. You tighten up your session timeout policy. And suddenly, no one can log in. “Did we test this in staging?” “Yes, kind of.” If you’re testing authentication like it’s just another API, you’re missing the bigger picture. Authentication flows are user-critical and security-sensitive, and they deserve a dedicated, realistic test strategy. 🔐 Here’s How I Test Auth Flows Without Touching Prod 1️⃣ Use a Mirror Staging Environment Duplicate your production auth settings in staging Same identity provider (e.g., Azure AD, Okta, Firebase Auth) Same token policies, same redirect URIs, but with test clients/apps 🎯 Goal: Simulate exactly how the real login behaves, without real user impact. 2️⃣ Simulate Real-World Scenarios, Not Just Happy Paths Too many teams test only the basics: Login succeeds ✅ User lands on dashboard ✅ But what about: Login from a new device? Expired token on page refresh? Session timeout and forced re-auth? Logging in via SSO with wrong permissions? 💡 Test the edges, that’s where things break silently. 3️⃣ Test Multiple Auth Flows Depending on your stack, you may have: 🔐 SSO via SAML or OIDC 🔓 Passwordless via email or biometrics 🔁 Token refresh via silent re-auth 📲 MFA with push, SMS, or TOTP Each of these flows behaves differently in test vs. prod and needs its own test case. 4️⃣ Use Fake Users with Real Roles Create sandbox users with roles like Admin, Viewer, Billing, Support Run access tests for each, especially privilege escalation attempts Try invalidating sessions mid-use 🧠 Auth is more than login, it’s what you can (and can’t) do once you’re in. 5️⃣ Monitor Logs & Alerts Even in Staging If your login fails in staging, but no one saw it, it’s as good as untested. Enable: Auth event logs Token grant/denial reports Audit trails for who accessed what (and when) 🧠 Bonus: Do a Dry Run With a Non-Prod Identity Provider Many platforms let you use a separate directory for dev/testing. If possible, isolate your test logins entirely and practice revoking access, triggering MFA, and simulating brute force attempts. 💬 Final Thought: Test your login flows like an attacker but protect your prod like your reputation depends on it. Because it does. How does your team test auth changes safely today, and what’s the step you used to forget? Let’s help each other build safer rollouts. #HowToFriday #AuthenticationTesting #DevOpsSupport #StagingBestPractices #LoginFlows #SSO #OIDC #MFA #TokenSecurity #UserAccess#ITSupport#Security
-
Shivany S.
Starbucks • 3K followers
A great resource for those looking to keep architecture diagrams inline with their source code in their Git Repo. I, personally, have liked both Mermaid Chart and PlantUML. But, looking to evaluate others mentioned in this list. It also helps tremendously if the tool has IDE plugins with Visual Studio Code or IntelliJ.
19
-
Karanbeer Singh - Certified BigCommerce and Shopify Developer
Cronix • 1K followers
Sidekick Upgraded: From Assistant to Operator 🤖 Shopify Sidekick evolves beyond reactive suggestions. In Winter ’26, Sidekick becomes proactive, surfacing high‑impact insights right when you or your merchant needs them — no more hunting for data. Inside Retail Australia Highlights: • Sidekick Pulse: Smart, personalized next steps based on store data. Inside Retail Australia • Generates custom internal tools and automation workflows from natural language prompts. Shopify • Reusable skills for repeatable tasks save time across projects. #shopify #ecommerce #shopifystore #dropshipping #shopifydropshipping #shopifyseller #smallbusiness #ecommercebusiness #shopifyexperts #digitalmarketing #fashion #entrepreneur #onlineshopping #marketing #business #amazon #etsy #shopifyexpert #onlinebusiness #wordpress #shopifybusiness #shopifypartners #shopifytips #shopsmall #shopifypicks #webdesign #shopping #onlinestore #shopifywebsite #shopifydeveloper
8
2 Comments -
Ehsan Gazar
Tipalti • 10K followers
Most engineers talk about architecture principles. Few actually build systems that follow them. I've seen this gap at Tipalti, at Mecca Brands, and in every codebase I've inherited. Teams write code that works today. Then traffic doubles. Or requirements change. Or someone leaves. And everything falls apart. The difference between systems that work in production vs. systems that work in demos? Five principles: 🔧 Scalability - Handle growth without rewriting everything 📊 Manageability - Debug and fix issues in minutes, not hours 🧩 Modularity - Change one part without breaking everything 🔌 Extensibility - Add features without touching existing code ✅ Testability - Deploy on Fridays with confidence I just wrote a deep dive on how to actually implement these principles, with real examples from systems I've built and systems I've fixed. Not theory. Practical trade-offs and decisions. Read it here: https://lnkd.in/eaAtpGrU Which principle have you ignored that came back to bite you? Share your war stories below 👇
2
-
Courtney Youngblutt
Iress • 3K followers
🤓Xplan Hint: Did you know you can set how data has to be entered into certain fields with a little Regex code? 🥱 Tired of seeing 'QLD', 'Qld' & 'Queensland' intermittently throughout your database? Frustrated with invalid email addresses or phone numbers with missing digits? Xplan has supported Regular Expressions (Regex) since I can remember. 😎 This underutilised but super cool functionality helps enforce data formats and improve data quality. and (with a little help from AI) - it's super easy! Andrew Wilson has provided some practical examples on how you can use Regular Expression (Regex) in Xplan in this latest Xplan Hint. https://lnkd.in/gSi7ikXA #xplan #xplanhint #advisely
6
-
Saransh Seth
Intrepid Travel • 576 followers
Most engineering orgs are solving the wrong problem with middle management. Gergely's latest Pulse covers the trend toward flatter teams and fewer middle managers. Having led Phoenix Design System across 3 product teams at Intrepid, I reckon the issue isn't the number of managers — it's what they're managing. In my experience, the best "management" happens when you remove the need for it. Phoenix works because the system itself enforces consistency. 23 components, clear patterns, self-documenting APIs. The design system IS the middle manager. Most teams add managers to solve coordination problems that good tooling and clear interfaces would eliminate. You don't need a PM to coordinate API changes if your API is well-designed. You don't need a manager to enforce code standards if your linter is unforgiving. The teams that thrive with fewer managers aren't the ones with superhuman ICs — they're the ones who've invested in systems that make coordination automatic. Like a well-designed game engine where the rules prevent chaos without a referee. Flexible teams work when the flexibility is constrained by good boundaries, not endless meetings. What coordination problems could your team solve with better systems instead of more managers?
-
Andrea Magnorsky
I am a driven, principled… • 3K followers
I went to ProductTank Auckland meetup "Owning the Financials - The Future of Product Management?" on the 27th of August I haven't been to a product meetup on this hemisphere and very happy I did. I've been meaning to post about this because I did learn a lot, great to go to meetup for adjacent roles, there is always more in common than one assumes. Corinna Stukan Shared an experience from a Swedish market where a single high‑value customer accounted for ~90 % of revenue (It was a high percent, could have been lower than this? corrections welcome 😅 ) Key thing I remember her saying: "Listening as a strategic skill—understand the real pain before building solutions. Anthony Marter Build strong relationships with finance early; treat finance as a partner rather than a gatekeeper. Forecasting and validating assumptions should be a continuous, collaborative effort. Key thing I remember him saying: If you don't know, make it up (educated guess) and talk to people to learn what is wrong with your guess Vivek Kumar Highlighted the “delta” between product vision and CEO expectations, especially around sales and revenue forecasts. Key thing I remember him saying: build something to test, and validate quickly. Originally posted in https://lnkd.in/eSXUMckY
15
-
Peter Wei
Redbourne Group • 4K followers
"Vibe coding." It sounds great, doesn't it? You just vibe. The AI writes the code. You build a whole app in an afternoon without understanding a single line of syntax 🤨 But Lee Simpson dropped a truth bomb at the recent Microsoft Azure User Group Brisbane that every dev needs to hear: "It's fun... until it's not." We've all been there. You start a project with Copilot or ChatGPT. It feels like magic. Then you hit the "Complexity Wall” Suddenly, the AI gets confused. It starts hallucinating. You try to stuff 300k tokens of context into the prompt, and the results get worse, not better. Lee calls this the difference between Coding and Engineering. Coding is just typing (AI is so great at typing!) However, Engineering is requirements, testing, and actually understanding the mess you just made. Three things I took away from his talk: 1️⃣Don't "stuff" the context: Just because the model accepts 1 million tokens doesn't mean it understands them. It gets tired, just like us. 2️⃣Fix the real bottleneck: Usually, it's not the coding speed. It's the messy requirements. AI can't fix bad instructions. 3️⃣The power of "Yet" : If AI can't do it today, add the word "yet." It moves too fast to write it off. Stop trying to replace your brain with a prompt. Use the tool to move faster, but don't forget how to drive 👨🏼💻
42
5 Comments -
khaled nabil
Intalio • 660 followers
Your title doesn't define your leadership. I'll share how to effectively champion significant technical initiatives and steer architectural direction without a 'Staff' or 'Principal' badge. Many of us senior individual contributors find ourselves driving critical technical vision and influencing architectural decisions without the formal 'lead' title. This isn't just about doing the work; it's about leading through influence, often from the middle of the engineering hierarchy. Here's how to effectively lead from the middle: 1. **Articulate the 'Why':** Don't just present a solution; clearly communicate the problem, the vision, and the long-term impact on the product and team. Connect your ideas to broader business objectives to gain buy-in. 2. **Build a Coalition:** Identify key stakeholders across teams and levels. Seek their input early, understand their concerns, and incorporate their feedback. Influence is built on trust, empathy, and shared understanding, not just mandate. 3. **Prototype & Prove:** Sometimes the best way to lead is to *show*, not just *tell*. Build a small proof-of-concept, gather data, and demonstrate the tangible benefits of your approach. Data-driven leadership is incredibly powerful and persuasive. 4. **Mentor & Empower:** Elevate others by guiding them through complex problems, sharing your expertise, and giving them ownership over parts of the solution. When your team thrives, your technical vision gains momentum and advocates. This approach not only drives critical technical work but also builds your reputation as a trusted, influential leader, regardless of where you sit on the org chart. It's about making a significant impact from any position. #TechLeadership #SoftwareEngineering #CareerGrowth #IndividualContributor #TechnicalVision #EngineeringLeadership #Influence What's your most impactful 'lead from the middle' story? Share your experiences below! #TechLeadership #SoftwareEngineering
-
Will Velida
Microsoft • 3K followers
Tomorrow I'll be speaking at DDD Melbourne for the first time, giving a talk about Platform Engineering in the age of GenAI. In this talk, we’ll explore how platform engineering and generative AI can work together to improve developer experience, if you get the foundations right. We’ll look at common organisational and cultural challenges that limit productivity, and how to start addressing them with simple, practical steps. From there, we’ll show how “everything-as-code” patterns and AI-assisted tooling can accelerate platform maturity and reduce friction across teams. Hope to see you there! #DDDMelb
51
3 Comments -
Dasith Wijesiriwardena
Microsoft • 2K followers
Come catch my session on Context Engineering at DDD Melbourne this Saturday. We've all been there: GitHub Copilot promises to be your coding companion, but instead feels more like that overeager intern who confidently writes brilliant code for the wrong problem. As AI-assisted development tools become ubiquitous, the industry narrative promises revolutionary productivity gains. Yet many practitioners find themselves playing an exhausting game of context whack-a-mole—constantly explaining, re-explaining, and fixing what their AI "partner" confidently got wrong. Having spent considerable time wrestling with this gap between AI promises and reality, I've discovered that the real challenge isn't prompt engineering—it's context engineering. This talk explores why "vibe coding" is fundamentally broken and introduces the systematic discipline that separates magical AI experiences from expensive disappointments. Through the constraint-context matrix and real examples like the Breadcrumb Protocol, I'll demonstrate why human expertise in scaffolding, steering, and domain understanding isn't just relevant—it's the secret ingredient that makes AI actually work. You'll leave with practical strategies for engineering context systematically and a framework for building sustainable human-AI collaboration that leverages both your skills and the machine's capabilities.
48
2 Comments
Explore collaborative articles
We’re unlocking community knowledge in a new way. Experts add insights directly into each article, started with the help of AI.
Explore More