Community Spotlight

Daniel Kimmelmann works on Postman Flows, where his work sits at the intersection of API-driven workflows, testing, automation, and enterprise reliability. His work reflects a broader shift across modern API environments as organizations are discovering that testing individual endpoints is no longer enough to validate how production systems actually behave.

Traditional API testing models were designed around isolated requests. Teams validated whether an endpoint returned the expected response, authenticated correctly, or matched a predefined schema. But enterprise systems increasingly operate through workflows that span multiple APIs, services, event streams, third-party platforms, and automation layers simultaneously. A single customer action may trigger dozens of interconnected API calls across authentication systems, payment services, messaging infrastructure, analytics pipelines, and internal business applications.

The challenge is that failures appear across sequences of interactions. An endpoint may pass contract testing while still breaking downstream workflows when deployed to production environments due to retries, latency variations, asynchronous processing, or dependency failures. APIs that behave correctly in isolation can still create instability when combined into larger execution paths.

Kimmelmann’s work centers on treating APIs as connected workflows. This shifts testing toward validating how systems behave together under production-like conditions. Authentication transitions, chained dependencies, event timing, retry amplification, rate limits, conditional logic, and cross-service execution are included in testing. Platform teams, product teams, security teams, and business operations all depend on the same workflows, even when they own different parts of the system.

Workflow automation and AI-connected systems execute requests continuously, trigger actions across multiple services, and dynamically chain operations together. Small inconsistencies that developers might manually work around can compound into larger failures once workflows become automated. Systems that appear stable under human traffic patterns may behave very differently under machine-driven execution.

The implication is that reliability depends on how well organizations test business workflows rather than individual technical components. Inside CI/CD pipelines, testing includes validating whether entire workflows continue to function predictably. A checkout flow, onboarding sequence, claims workflow, or support escalation path may depend on multiple APIs owned by separate teams. Testing those paths as complete workflows gives enterprises a clear view of operational risk before customers, employees, or automated agents encounter it in production. The practical question shifts from whether an API works to whether the larger workflow built on multiple APIs continues to operate reliably under real-world production conditions.

API Feed

Big Story

For years, enterprise APIs were designed with a relatively clear assumption that the end consumer would be another human developer. Documentation explained how endpoints worked, engineers interpreted edge cases during integration, and operational inconsistencies were often resolved through support tickets, tribal knowledge, or direct communication between teams.

Modern enterprise systems are consumed by a mix of humans, automated workflows, internal platforms, external integrations, and AI agents that operate simultaneously on the same infrastructure. APIs sit at the center of systems where requests may originate from developers, event-driven pipelines, copilots, orchestration layers, autonomous agents, or machine-generated workflows.

The consequence is that APIs designed primarily for human interpretation often behave poorly under machine-driven execution.

Human developers are generally adaptable. They can infer undocumented behavior, work around inconsistent naming conventions, manually retry failed integrations, or interpret ambiguous error responses. Automated systems and AI agents are significantly less tolerant. They depend on predictable contracts, structured responses, stable authentication behavior, explicit execution rules, and reliable workflow semantics to operate safely at scale.

This changes what good API design means in practice.

Historically, many enterprise API programs focused on quickly exposing functionality such as connecting systems, enabling integrations, and supporting application development across teams. The new challenge is interoperability among multiple consumer types simultaneously. APIs need to function as stable execution layers for humans, systems, and agents operating under very different assumptions and failure patterns.

The pressure is growing as organizations expand AI-connected workflows. Agents do not consume APIs the same way humans do. They chain endpoints dynamically, aggressively retry operations, execute workflows continuously, and interact across systems that were not originally designed to work together autonomously. Small inconsistencies that a developer might ignore can create cascading issues when multiplied across automated execution paths.

This is pushing platform teams toward more structured API design practices. Machine-readable contracts, standardized authentication flows, explicit rate-limit behavior, deterministic error responses, workflow-aware documentation, and stronger schema consistency are key operational requirements.

The shift also affects observability and governance. APIs can no longer be evaluated only through uptime or request volume. Teams need visibility into workflow completion behavior, agent execution paths, retry amplification, dependency chains, and cross-system interactions. Large enterprises face additional complexity because APIs often evolve unevenly across departments, acquisitions, legacy systems, cloud environments, and third-party platforms. Different authentication models, inconsistent documentation standards, fragmented ownership, and overlapping integration patterns create friction that becomes harder to manage as machine-driven consumption increases.

The broader transition is that APIs are evolving from integration layers to coordination layers for enterprise uses. As organizations expand automation and AI-driven systems, the APIs that succeed will be those designed not only for human usability but also for predictable execution across every type of consumer interacting with the enterprise stack.

Resources & Events

apidays Amsterdam brings together API practitioners, architects, and platform leaders for two days focused on operating APIs as products in enterprise environments. The 2026 program includes sessions on API design, lifecycle management, governance, and platform strategy, as well as discussions on how APIs are evolving alongside AI-driven systems and automation. The event is designed for teams building and scaling API programs across organizations, with a mix of technical sessions and practitioner-led case studies. Details →

apidays Munich brings together API architects, platform engineers, and enterprise technology leaders for two days focused on API strategy, platform operations, and AI-driven enterprise systems. The 2026 program includes sessions on API lifecycle management, developer experience, event-driven architectures, and the operational challenges of managing APIs across distributed environments. Designed for teams scaling enterprise API programs, the event combines technical discussions, practitioner-led case studies, and platform engineering perspectives on how APIs are evolving alongside automation and AI-connected workflows. Details →

You can find a list of all Apidays events here

Apply to speak at Apidays Singapore, NY, London, Paris, and more here

Platform Engineering Week brings together sessions and workshops across Platform Engineering Summit, DevOpsCon, MLCon, and API Conference. The program focuses on internal developer platforms, automation, governance-by-default, AI-ready delivery pipelines, API contracts, platform boundaries, LLMOps, and production deployment patterns for engineering, DevOps, SRE, API, and AI platform teams. Details →

Mia-Platform’s report examines how internal developer platforms are evolving as AI becomes part of software delivery workflows. Based on survey responses from engineering and platform professionals, the report examines platform maturity, developer self-service, governance, observability, and the operational challenges posed by growing architectural complexity. One recurring theme is that many organizations are still building foundational platform capabilities while simultaneously trying to integrate AI into development, support, automation, and operational workflows. The findings also show growing pressure on platform teams to balance speed, standardization, security, and developer autonomy as engineering environments become more distributed and toolchains become harder to manage. Read →

Insight of the Week

The more steps an autonomous workflow contains, the faster failure rates compound across the system. Research on multi-step agentic workflows found that even workflows with 90% reliability at each individual step complete successfully only about 35% of the time across ten-step execution chains. The implication is significant for enterprise API and automation teams building AI-connected workflows. Reliability problems are shifting from individual model performance toward orchestration, workflow verification, retry management, and execution controls across distributed systems.

For the Commute

This session looks at how AI is putting pressure on energy use, water consumption, hardware demand, and infrastructure planning in ways many organizations are not prepared for. The speaker, Lisa Pratico, argues that enterprise architecture teams have often treated sustainability as a reporting issue rather than a system design problem, even as AI workloads place demands on compute, cooling, and operational infrastructure. Much of the discussion focuses on practical changes such as right-sizing workloads, relocating compute to lower-impact regions, reducing idle infrastructure, and measuring the carbon impact of application portfolios alongside cost and business value.

Stay tuned for bold ideas, fresh perspectives, and the next wave of API innovation

-The Apidays Team