Observability, Product Intelligence, and AI Agents Converge
The latest wave of developer tools funding highlights a convergence of observability platforms, product intelligence tools, and AI agent platforms that all aim to give engineering teams deeper, machine-readable context about how software behaves, how work gets done, and where risk is emerging across complex systems. Undo, Devplan, and Tsuga together show how investors are backing platforms that connect runtime telemetry, organizational knowledge graphs, and AI-native monitoring so both humans and agents can understand, debug, and guide modern applications. Their combined raises of about USD 74.5 million (approx. RM343.7 million) signal that observability and product intelligence are no longer separate categories but complementary layers in an emerging stack built for AI-driven development. This stack is designed to keep pace with accelerated code generation, autonomous deployments, and dense interaction logs that traditional tooling struggles to handle at scale.
Undo Turns Runtime Context into an AI-Friendly Debugging Layer
Undo’s USD 37 million (approx. RM170.2 million) growth investment led by Elsewhere Partners underscores investor belief that debugging is becoming a data problem as much as a workflow problem. Undo records deterministic runtime traces so engineering teams and AI coding agents can see what actually happened inside a system instead of relying only on static analysis. According to Undo, AI agents solve 38% of complex bugs with static code alone, but that figure rises to 92% when they have access to runtime context. The company says automated root-cause analysis can make mean time to resolution 100 times faster for complex incidents. With new capital, Undo plans to deepen integrations into AI engineering workflows, expand customer success, and embed its observability platform more tightly into agentic engineering pipelines that support some of the world’s most complex codebases.
Devplan Builds a Product Intelligence Layer for Coordination at Scale
At the earlier stage, Devplan’s USD 2.5 million (approx. RM11.5 million) seed round, led by AI2 Incubator and Acequia Capital, targets a different bottleneck: shared understanding. Co-founder and CEO Chris Bee argues that “the bottleneck is no longer building software. It’s maintaining shared understanding.” Devplan connects tools like Slack, Jira, GitHub, documentation systems, and meetings into a unified product intelligence layer. At its core is Weaver, a knowledge graph that continuously tracks what changed, why it matters, and what needs attention. This structure gives product and engineering leaders real-time visibility into progress, risk, and decisions while also feeding AI agents with the same organizational context humans rely on. Early users report reclaiming 8–10 hours per week and seeing context queries run twice as fast and 3.5 times more cheaply than AI workflows that query source systems directly.
Tsuga Targets AI-Native Observability with an In-Cloud Architecture
Tsuga’s €30 million Series A, equal to about USD 35 million (approx. RM161.3 million), reflects a strong bet on AI-native observability architecture. Instead of shipping telemetry to a third-party cloud, Tsuga deploys inside each customer’s own cloud accounts across major hyperscalers and regional sovereign environments. This keeps telemetry under customer control, removes infrastructure duplication, and allows AI models to run on complete, unsampled data. The company argues that legacy observability platforms were built on an assumption that the AI era has invalidated: that it is reasonable to ingest and store everything centrally as volumes grow. Every agent loop and autonomous deployment now generates telemetry at levels that make traditional ingestion-cost models harder to sustain. Tsuga prices on a single rate per gigabyte, with costs expected to fall over time as engineers tune each deployment, and already reports several millions in revenue with six-figure average contracts.
A Funding Pattern for the Next Developer Stack
Viewed together, these three software startup funding rounds show a clear pattern in developer tools funding: investors are backing both mature observability platforms and early-stage product intelligence tools at the same time. Undo’s growth round validates a market for runtime-aware debugging that plugs directly into AI coding agents. Tsuga’s Series A shows appetite for AI-native observability platforms that run inside customer clouds and are built for dense agent traffic. Devplan’s seed round fills the coordination gap with a product intelligence layer that organizes human and machine context across the development lifecycle. Their combined capital—about USD 74.5 million (approx. RM343.7 million)—signals that future engineering stacks will blend observability data, organizational knowledge graphs, and AI agent platforms into a single fabric. In this model, AI systems do not sit on the side; they become first-class consumers and producers of context for building, operating, and improving software.
