Today: Microsoft rolled out its second wave of Copilot feature upgrades ahead of a pivotal year for its AI strategy, AWS throws Intel a lifeline, and the latest funding rounds in enterprise tech.
Today: Salesforce jumps on an AI bandwagon with its trademark enthusiasm, Apple reveals more details about its secure private cloud infrastructure, and the latest enterprise moves.
For years, Oracle tried to convince longtime database customers who wanted to shed their on-premises data centers to run those databases on Oracle's public infrastructure cloud, slamming AWS at every turn. Times have changed.
DevOps walked so platform engineering could run, but building a standardized organizational approach to software development can cause as many problems as it aims to solve if the platform's foundation is shaky. Here's how eight experts think companies should approach platform engineering.
DevOps walked so platform engineering could run, but building a standardized organizational approach to software development can cause as many problems as it aims to solve if the platform's foundation is shaky. Here's how eight experts think companies should approach platform engineering.
SVP & Head of Engineering and Technical Support, Wrike
To maximize the ROI from platform engineering, organizations need a long-term vision. Without it, they won’t design the platform correctly and can make it too abstract, hindering feature development. Organizations also require a clear product strategy for the next 12-18 months. Without this, platform and product development will be misaligned.
Organizations need talented developers and a qualified architect to design the platform, because building a platform is more complex than feature development. Implementing the right SDLC is also crucial for success. While dev teams should follow agile development principles, the platform core's architecture should be designed simultaneously with extensibility, scalability, and product vision in mind from the start. Platform features can be designed later, but an improperly designed platform core will slow future development instead of speeding it up.
Once the platform core’s beta version is ready, organizations can control ROI by ensuring feature teams develop faster on the platform, measured via team velocity. They should gather feedback from dev teams using the platform to understand its value and ensure the platform is aligned with the product.
Senior vice president, developer platform, Oracle Cloud Infrastructure
Organizations that embrace platform engineering share two characteristics in their tools and infrastructure choices: they are homogenous and opinionated. They’re using solutions built on open standards with additional fine-tuning to incorporate parameters to suit their development needs.
Kubernetes has become an essential tool for platform engineering teams, and is supported by an active community of contributors, including open-source project maintainers, commercial vendors, and corporate end users. More recently, Kubernetes has become a common solution to the challenges of AI model training and inferencing. Built-in autoscaling and self-healing capabilities help ensure models avoid downtime and can handle the required load in order to provide accurate predictions.
However, day-to-day maintenance can still take a lot of time, giving rise to fully managed Kubernetes services. These services enable teams to set further workload parameters to automate resource scaling. No longer will they have to worry about overprovisioning workloads. Everything will run to the scale needed. That’s when platform engineering becomes a strategic advantage, enabling DevOps workloads to be less expensive, and more transportable.
Partner, Stage 2 Capital
As an emerging category of software, platform engineering is not just about the software or vendors you work with, but also about clearly defining the role of platform engineering within your organization. Maximizing investment here includes clearly defining its purpose in improving developer experience, but it also requires getting buy-in from different parts of an organization.
Companies that are successful in implementing platform engineering efforts tend to view it less as an extension of their DevOps or SRE teams but rather as a distinct initiative within their company where developers, operations, and other stakeholders are involved in improving developer experience so their end users will ultimately benefit. No software solution, whether internally built or externally purchased, will succeed if there is not the proper buy in across the board.
VP, Digital Consulting, Ensono
To get the most out of your platform engineering investment, focus on automation. Modern software should be automated, not manually operated. Use scripts to build entire environments and tools like Terratest to automate testing for both infrastructure and applications, eliminating guesswork from provisioning.
Pay attention to cross-cutting concerns – the essentials every application needs. For instance, we spend a lot of time automating build pipelines with various test hooks to ensure consistent, repeatable software releases that don't introduce regressions. A well-designed, modular build pipeline can easily transfer from one application to another. The same goes for logging, monitoring, and alerting – platform engineering should standardize these tools to ensure consistent instrumentation across all applications, as they'll likely be monitored on the same APM or SIEM dashboards.
Invest in governance, too. Establish processes and procedures to manage platform engineering investments, avoiding siloed work and duplicated efforts. Platform engineering is an ongoing process – always plan for continuous improvement.
CMO and Developer Advocate, SingleStore
A lot of platform engineering tasks are spent on non-platform engineering work, such as responding to tickets and reviewing PRs, which takes valuable time away from platform engineering. I have seen a number of companies focus on eliminating this using automation.
With the advent of gen AI technologies, we have started to see agentic automation tools that can automate tasks like classifying, prioritizing and assigning tickets, crafting responses and even taking actions that are repetitive and not directly related to system uptimes. Companies that are dedicated to automation and connecting the entire pipeline of observability, monitoring, incident reporting, resolution and reporting are the ones that are becoming more agile and able to focus on moving fast as entire organizations.
SVP of Infrastructure and Product Operations, Netskope
Today's platform is the common foundation supporting internal infrastructure, products, services, and customer interactions. A nuanced approach accommodates diverse workloads and capacity models, facilitating a zero-downtime expectation. Recent global events underscored the importance of vendor independence, enabling us to navigate supply chain challenges without constraining growth.
Close relationships with integration partners allow us to leverage robust testing environments while pushing us to design an unwavering commitment to standardization. This leads to quick auto-remediation and simulated outage scenarios. A relentless focus on dependency mapping accurately represents the blast radius of every platform component. Finally, targeted AI investments integrate vastly different capacity models into a unified scaling model, allowing us to predictably scale ahead of demand.
SVP of Products and Engineering, Hypermode
Treat the platform like a product, complete with core metrics and a roadmap. The best platform teams orient around removing developer pain in shipping new features, while also increasing organizational outcomes of availability and security. There are a lot of stakeholders that care about these outcomes and the pressure can grow on young platform teams, especially after the first major outage. Taking a product mindset keeps the focus on the outcomes with clear priorities so that platform engineers can focus on development.
Great products also take accountability for the success of their users. If a developer isn't configuring caching in their build pipeline, as an example, it's the platform team's responsibility to help that developer find success. I've seen many debates emerge on who is at "fault" in this scenario. Orienting towards product outcomes ensures someone helps that developer find success with the platform. That's the goal, after all.
CTO, Veriff
Between good, fast and cheap, you've been told you could only pick two. Platform engineering is the uncompromising art of picking all three.
Before delving into maximizing its impact, let us first review the two ways to minimize it. First, don't size this investment based on industry benchmarks. You should be thinking about the leverage you want your engineering to have versus the one it currently has. The second pitfall I call dependency-driven platform engineering. Dependencies mean hard deadlines that involve compromises, but we have already established that they are incompatible with the mission at hand.
What then is needed to win? Two keywords: industrialization and enablement. The latter ensures an aligned concept of value between platform and product engineering, and here product and agile methodologies are relevant. Industrialization, on the other hand, is deep, hardcore engineering, the same mindset used to build bridges. Agile isn't the way for that.
Combining these two approaches is what maximizes the impact of platform engineering. I sometimes refer to it as making the right thing to do the path of least resistance.
Every company understands the value of their corporate data, but it's easy to lose track of priorities when trying to update their toolsets, especially during the generative AI frenzy. Here's how eight experts think companies should navigate the tricky road to the modern data stack.
Despite some recent gains by law enforcement, ransomware remains a pernicious problem for companies large and small. Here's how eight experts advise preparing for ransomware attacks.
