How to Transform Development Workflows

Engineering transformation is not optional anymore, it’s a race against irrelevance! For years, we’ve all seen the same patterns in product development. - Mechanical, E/E and software teams working in isolation. - Complexity growing faster than our ability to manage it. - Errors discovered too late. - Interfaces that don’t fit. Integration often feels like assembling a puzzle… only to realize that half the pieces were built from entirely different pictures. Weeks, sometimes months, lost not because of bad engineering, but because of fragmented engineering. And yet, despite knowing these problems for years, many organizations are still waiting. Waiting for the “right moment.” Waiting for clearer standards. Waiting for others to move first. That moment is gone. Global competitors have already picked up speed and are exerting pressure. With Model-Based Systems Engineering (MBSE) and AI reaching real maturity, we finally have the tools to fix what we’ve been complaining about for a decade. The question is no longer if transformation will happen. The question is: how fast can you move? Here’s how Vlad and I currently think about it in 9 concrete steps: 1. Adopt and mature MBSE - Build system models that truly reflect your product, not just documentation. 2. Derive domain-specific models from system models - Create consistent, hierarchical product structures across all domains and disciplines. 3. Capture all engineering artifacts From requirements (RFLP) over testing to homologation, make everything explicit and create development templates. 4. Link all artifacts via a knowledge graph - Enable impact chain analysis based on a solid engineering ontology. 5. Standardize and accelerate component development - Align tools, data and processes for each discipline and component 6. Build cross-domain CI/CD pipelines - Enable fast, automated iteration across requirements, architecture, design, simulation and testing. 7. Rationalize the toolchain (APIs over UIs) - Tools must be controllable from the outside enabling agent-based workflows. 8. Make engineering knowledge machine-readable - Document not just the what, but the how and why. Only then can agents effectively navigate engineering-specific challenges. 9. Define the future work split - Clarify what engineers do and what AI agents should handle. Establish strong human-in-the-loop validation. The core message is simple: Engineering excellence in the future will not come from better tools alone. It will come from how well we connect systems, data, people and agents. Companies that start building this foundation now will gain speed. Those who wait will struggle to catch up. What’s missing from your perspective? Which steps would you add to make this transformation truly work? Timmo Sturm | Daniel Spiess | Sebastian Linzmair | Sascha Bach | Rick Bouter

Recently helped a client cut their AI development time by 40%. Here’s the exact process we followed to streamline their workflows. Step 1: Optimized model selection using a Pareto Frontier. We built a custom Pareto Frontier to balance accuracy and compute costs across multiple models. This allowed us to select models that were not only accurate but also computationally efficient, reducing training times by 25%. Step 2: Implemented data versioning with DVC. By introducing Data Version Control (DVC), we ensured consistent data pipelines and reproducibility. This eliminated data drift issues, enabling faster iteration and minimizing rollback times during model tuning. Step 3: Deployed a microservices architecture with Kubernetes. We containerized AI services and deployed them using Kubernetes, enabling auto-scaling and fault tolerance. This architecture allowed for parallel processing of tasks, significantly reducing the time spent on inference workloads. The result? A 40% reduction in development time, along with a 30% increase in overall model performance. Why does this matter? Because in AI, every second counts. Streamlining workflows isn’t just about speed—it’s about delivering superior results faster. If your AI projects are hitting bottlenecks, ask yourself: Are you leveraging the right tools and architectures to optimize both speed and performance?

From Code Generation to System Integration: Why AI Coding Tools and Agentic IDEs Must Evolve to Solve Real Software Development Challenges Since GPT-3 went mainstream, AI coding tools have sprinted through three waves. 1. First came smart autocomplete. 2. Then came cloud companions tuned to specific stacks. 3. Now we’re in the agent wave – tools that read whole repos, open terminals, run tests and raise pull requests on their own. Every cycle starts the same way: Wow. Impressive. Look at how much this can do for me. But the uncomfortable truth is this: most of what these tools automate is commodity knowledge. Framework boilerplate, CRUD patterns, standard integration glue, typical test shapes – once a pattern exists in public code, a model can learn it and repeat it very well. That used to feel like expertise. Now it’s autocomplete on steroids. The real problems have barely moved: • Design and architecture. Not just file-by-file edits, but coherent system design: boundaries, contracts, data flows, failure modes, performance budgets – a holistic solution, not local patchwork. •  End-to-end SDLC integration. How change actually flows from idea to production: design, review, CI, approvals, environments, rollout strategies and on-call ownership. • Change management and legacy transformation. How to evolve decade-old systems, untangle hidden dependencies, migrate behaviour safely and avoid breaking everything that still quietly depends on “that old module”. • Traceability. Knowing who or what changed what, why, and what else was impacted – across code, configs, data pipelines and policies. • How strongly workflows enforce the top 10 principles like reliability, security, cost and maintainability that were outlined in the earlier post – not as posters on a wall, but as gates every change must pass through. This is where vibe-coding tools become dangerous. The model writes the feature, generates the tests, explains the diff. Everything looks green. It feels safe enough to ship on vibe. Without deep expertise and a solid workflow around it, that is not productivity. It is an efficient way to inject new risk into a live system. If code patterns are now cheap, differentiation shifts somewhere else: • To how clearly an organisation defines how systems should be built and evolved • To how tightly AI tools are integrated with that SDLC, not just with the editor • To how well workflows embody design principles, change discipline and traceability by default Writing code is becoming a commodity. However, writing holistic, thoughtful systems, and continuously evolving and governing them safely, is where the true value lies AI coding copilots and agentic IDEs now need to evolve from “look what I can generate” to “look how I help you integrate, operate and transform”. That’s when it stops being “wow, impressive demo” and becomes “yes – this is finally solving the real problem.”

Digital transformation sounds exciting on slides. AI. Automation. Paperless. Everyone’s in a rush to “transform”. But the real work? It’s messy. I’ve seen projects stall halfway, budgets balloon or worse, teams quietly revert to old ways. Why? Because most digital transformation projects don’t start with clarity, they start with tools. Before you automate, digitise, or implement anything, untangle the mess first. Here are 5 things to do before you begin: ✅ 1. Map Reality, Not Assumptions Don’t rely on what managers think the process is. Follow the paperwork. Sit beside staff. Observe who’s emailing, printing, signing, chasing. Often, what’s in the SOP and what happens on the ground are two different worlds. Value tip: Do a shadowing session. Ask: “What’s slowing you down?” Not “What system do you want?” ✅ 2. Identify Invisible Dependencies Some approvals only happen over WhatsApp. Some people are unofficial gatekeepers. These aren’t in org charts, but they’ll derail your project if ignored. Value tip: Interview informal influencers. Understand who really moves things forward — or blocks them. ✅ 3. Decide What to Kill, Not Just What to Build Digital doesn’t mean copy-paste old ways into new tech. Don’t digitise junk. Clean it up. Kill redundant steps, outdated forms, duplicated approvals. Value tip: Run a “Keep / Kill / Automate” workshop with department heads. ✅ 4. Align the Pain With the People Transformation must solve real pain. Not just C-level goals. If end-users don’t feel the benefit, they’ll resist quietly or find shortcuts. Value tip: Prioritise based on frontline frustration. Not only what looks impressive in reports. ✅ 5. Commit to Iteration, Not Perfection Don’t spend a year building the perfect system. Go live with the 70% that works. Improve with feedback. Perfect kills momentum. Small wins build trust. Value tip: Pick one team or process. Prove it works. Then scale. . . . Digital transformation isn’t about software. It’s about unblocking humans. Start there, and your project won’t just go live. It’ll stick. Want help mapping or untangling your current workflow before you invest in tools? DM me. Let’s make the invisible visible.

How AI transforms entire workflows beyond incremental gains: I've been following the conversation at WEF Davos about AI adoption, and Andrew Ng's recent letter captures something crucial that many organizations miss: The thousand flowers approach doesn't bloom into transformation. Bottom-up AI experiments are valuable. Teams close to problems often see solutions first. But the real transformation requires redesigning workflows end-to-end, not just automating individual steps. Consider this: A bank automates loan approval from 1 hour to 10 minutes. That's nice efficiency. But it's transformative only when you ask: What changes when customers get decisions instantly? The answer: Everything. - Marketing shifts to highlight "10-minute decisions" - Applications need digitization and smart routing - Final review scales for higher volume - The entire product becomes competitive differently This is where strategy meets technology. AI isn't just a tool for efficiency. It's a catalyst for rethinking how work flows through your organization. Organizations creating real impact with AI aren't running isolated experiments. They're asking: How does this change our entire customer experience? How do we redesign workflows end-to-end? That's where incremental becomes transformative.

→ 𝐓𝐡𝐞 𝐈𝐧𝐯𝐢𝐬𝐢𝐛𝐥𝐞 𝐅𝐨𝐫𝐜𝐞 𝐀𝐜𝐜𝐞𝐥𝐞𝐫𝐚𝐭𝐢𝐧𝐠 𝐘𝐨𝐮𝐫 𝐏𝐫𝐨𝐣𝐞𝐜𝐭 𝐃𝐞𝐥𝐢𝐯𝐞𝐫𝐲: 𝐃𝐞𝐯𝐎𝐩𝐬 𝐇𝐚𝐯𝐞 𝐲𝐨𝐮 𝐞𝐯𝐞𝐫 𝐰𝐨𝐧𝐝𝐞𝐫𝐞𝐝 𝐰𝐡𝐲 𝐬𝐨𝐦𝐞 𝐭𝐞𝐚𝐦𝐬 𝐚𝐥𝐰𝐚𝐲𝐬 𝐝𝐞𝐥𝐢𝐯𝐞𝐫 𝐟𝐚𝐬𝐭𝐞𝐫 𝐰𝐡𝐢𝐥𝐞 𝐨𝐭𝐡𝐞𝐫𝐬 𝐥𝐚𝐠? 𝐓𝐡𝐞 𝐬𝐞𝐜𝐫𝐞𝐭 𝐨𝐟𝐭𝐞𝐧 𝐥𝐢𝐞𝐬 𝐢𝐧 𝐡𝐨𝐰 𝐃𝐞𝐯𝐎𝐩𝐬 𝐭𝐫𝐚𝐧𝐬𝐟𝐨𝐫𝐦𝐬 𝐩𝐫𝐨𝐣𝐞𝐜𝐭 𝐝𝐞𝐥𝐢𝐯𝐞𝐫𝐲, 𝐞𝐬𝐩𝐞𝐜𝐢𝐚𝐥𝐥𝐲 𝐢𝐧 𝐝𝐚𝐭𝐚-𝐝𝐫𝐢𝐯𝐞𝐧 𝐞𝐧𝐯𝐢𝐫𝐨𝐧𝐦𝐞𝐧𝐭𝐬. 𝐇𝐞𝐫𝐞’𝐬 𝐡𝐨𝐰: • Automated Data Pipelines Continuous ingestion, processing, and validation happen without manual effort, speeding up data readiness. • Real-time Monitoring Instant insights into data quality and flow mean issues are caught and fixed immediately. • Data Version Control Tracking dataset versions alongside code ensures your data stays reliable and reproducible. • CI/CD for Data Models Automated deployment of machine learning models accelerates innovation and updates. • Infrastructure as Code Provisioning scalable environments quickly and reliably removes bottlenecks. • Security & Compliance Automation Policies are enforced automatically - audits become smoother and faster. • Collaborative DataOps Culture When teams work together seamlessly, data delivery becomes faster and more efficient. • Cloud-Native Data Tools Cloud services give you on-demand scaling and rapid workflow deployment. • Automated Data Testing Early detection of data issues prevents costly delays downstream. • Observability & Analytics Feedback Performance insights guide continuous improvement in pipelines. • AI-Driven Data Orchestration AI optimizes complex workflows, saving time and reducing errors. • Edge Data Processing Real-time, low-latency processing at the edge transforms customer experiences. DevOps is not just a methodology - it’s a supercharger for your project timelines. It turns complex, slow workflows into streamlined engines of innovation. Follow Carlos Shoji for more insights

The rise of AI Agents has transformed coding in just 3 years Here's the evolution most leaders are completely missing... If your team is still manually writing every line of code, you're already behind. The coding landscape has shifted from Traditional → Vibe → AI-Assisted → Agentic, and each stage requires a different mindset. 📌 Let me break down when to use each approach: 1/ Traditional Coding - Writing code manually line-by-line in a programming language.  - You build PRDs, write syntax, compile/interpret, debug errors, test for issues, then deploy. - Use: When you need full control, custom logic, or complex architecture that AI can't handle yet. Tools: VsCode, IntelliJ, Sublime Text Best for: Production systems where every line matters and security is critical. 2/ Vibe Coding - Describe what you want in plain language and let AI generate the entire app.  - Choose the right tool, write a query in natural language, let the LLM build your idea, add tools and databases, get feedback, then test and deploy. - Use: When you need quick prototypes, simple apps, or you're learning new frameworks. Tools: Bolt.new, Lovable, Replit Agent Best for: MVPs, landing pages, or internal tools where speed beats perfection. 3/ AI-Assisted Coding - You write code while AI suggests completions, like having a senior dev pair-programming with you.  - Build PRDs, developer verifies code, AI shares suggestions, you run debugging, write test cases, and maintain compliance. - Use: When you need production-grade projects requiring oversight but want 3x speed. Tools: Github Copilot, Code Whisperer Best for: Enterprise applications where human review is mandatory. 4/ Agentic Coding - AI agents autonomously code in iterative loops, building plans, writing code, fixing errors, checking test cases, and deploying with minimal human intervention. - Use: When you need complex workflows or end-to-end automation but you are willing to spend time reviewing the entire code. Tools: Claude Code, OpenAI Codex Best for: Automating repetitive tasks, batch processing, or multi-step workflows. The biggest mistake I see? Teams trying to use the same approach for everything. Traditional coding for a quick prototype? You'll waste days. Agentic coding for mission-critical banking software? A disaster. Here's the truth: The best teams in 2025 aren't the ones who code the fastest; they're the ones who know which method to use when. Master this evolution, and you'll 10x your output while others debate whether AI will replace them. 📌 If you want to understand AI agent concepts deeper, my free newsletter breaks down everything you need to know: https://lnkd.in/g5-QgaX4 Save 💾 ➞ React 👍 ➞ Share ♻️ & follow for everything related to AI Agents

With Skills, Hooks, AGENTS.md, and frontier models like Claude Opus 4.5 and GPT-5.4 in Codex, software development is changing fast. The job is no longer just typing code line by line. It is engineering the workflow around coding agents. That means: defining intent clearly writing better specs giving the agent the right context designing test scenarios early reviewing output with judgment and refining skills and instructions so the system improves over time The best developers will not be the ones who type the fastest. They will be the ones who can: think clearly guide agents well spot weak output quickly and keep a strong human feedback loop My view: Human-in-the-loop is still the biggest force multiplier in today’s developer workflow. Agents can generate. But humans still provide context, taste, accountability, and final judgment. Most engineering leaders I talk to are already budgeting for this shift: time to figure the right workflow that works for individuals and teams, AI tokens, and new team habits. We are moving from line-by-line coding to intent-driven software development. A simple workflow that works well for me today: Intent → Spec → Context → Agent → Tests → Review → Refine This has changed how I approach software development. How are you adapting your workflow for coding agents? #AIEngineering #SoftwareDevelopment #DeveloperProductivity