What Is Tsuga? Funding, Founders, and How Its AI-Native Observability Platform Stacks Up Against Datadog and Dynatrace

Tsuga emerged from stealth with the seed round in November 2025 and announced the Series A barely six months later, in June 2026. The speed of the follow-on round signals strong investor belief that the company's thesis — that observability must be rebuilt for AI-native, BYOC infrastructure — is resonating.

What Distinguishes Tsuga from Datadog and Dynatrace?

Tsuga's differentiation from incumbents like Datadog and Dynatrace centers on three axes: deployment model, pricing architecture, and AI-native design.

Bring Your Own Cloud (BYOC) Deployment

Instead of ingesting telemetry data into a vendor-controlled SaaS platform — the model Datadog and Dynatrace both use — Tsuga runs entirely inside the customer's own cloud environment. The Tsuga platform can be deployed across Microsoft Azure, AWS, Google Cloud, and sovereign cloud infrastructures, ensuring telemetry data never leaves customer control. This is a meaningful differentiator for organizations in regulated industries or with strict data sovereignty requirements.

One detail worth noting: while both Datadog and Dynatrace do offer some deployment flexibility options (Dynatrace offers managed and SaaS deployments; Datadog is primarily SaaS), the sources provided do not directly substantiate a claim that either incumbent is uniformly less flexible for data-sovereignty needs. Tsuga's BYOC-first architecture is clearly a bet that sovereignty and control will become increasingly important as AI agents generate more sensitive operational data.

Pricing Model: Challenging Per-Byte Pricing

Tsuga has explicitly positioned itself against the per-byte pricing model that Datadog helped popularize. As AI workloads make telemetry volumes explode, the argument goes, per-byte pricing becomes unsustainable for enterprises running large-scale AI agent fleets. Tsuga's pricing model is structured to decouple cost from data volume.

The sources do not provide exact pricing figures for Tsuga, and they also do not support a blanket claim that both Datadog and Dynatrace always price strictly per byte in all plans. Dynatrace, for example, has historically offered host-based pricing, while Datadog's pricing varies by product. The core claim that Tsuga is pricing against volume-based models is well-supported.

AI-Native and Agent-Observable Architecture

Tsuga is described as "observability software for the age of AI agents." Its platform is built so that AI agents can consume observability data directly. According to Tsuga's product documentation, the storage and query layer is designed to handle the data volumes that AI agents actually generate, and the APIs return "relevant context rather than raw data dumps, so agents spend their tokens on reasoning rather than on filtering noise."

This is a notable contrast with incumbent platforms, which the sources suggest are not natively designed for session-level AI agent tracing. A comparison analysis from Sentrial explicitly notes that neither Datadog nor Dynatrace offers native session-level agent tracing — both require custom instrumentation.

Tsuga also offers an "Agent-Native Observability" solution that lets engineering teams building AI agents connect observability data to every data source in their environment, not just the integrations a third-party platform has chosen to support.

How Does Tsuga Handle AI Agent Observability?

Tsuga's agent-native observability is built on three design principles:

1. Cost-efficient storage and query: The platform handles the data volumes that AI agents actually generate, rather than charging based on raw ingestion volume.
2. Context-aware APIs: Instead of returning raw telemetry dumps, Tsuga's APIs return relevant context, so AI agents can spend their compute budget on reasoning rather than filtering noise.
3. Full-environment connectivity: Because the platform runs in the customer's own cloud, agents can connect to every data source in the environment — not just the ones a third-party platform has chosen to integrate with.

Tsuga's Series A announcement positions the company as "the leader in AI-Native Resilient Observability" and states that the platform is designed to power a new generation of AI agents.

What Traction Has Tsuga Achieved Since Emerging from Stealth?

Since coming out of stealth in November 2025, Tsuga has achieved meaningful early traction:

Revenue and scale: Multiple sources report that Tsuga has "several millions in revenue" with six-figure average contract values. However, these revenue figures are company self-reports and have not been independently verified by the provided sources. Tsuga processes tens of terabytes of telemetry data every day running on AWS.

Customer base: Verified customers include:

• Le Monde, one of France's leading media groups, which deployed Tsuga under the leadership of its CTO/CIO Paul Laleu
• Black Forest Labs (a frontier AI model lab)
• Camunda
• Buk

Results at Le Monde: According to an AWS case study published in June 2026, within three months of deploying Tsuga, Le Monde achieved:

• A 50% increase in observability adoption among development teams
• A threefold expansion in monitored services and applications
• A 30% reduction in mean time to detect incidents (MTTD)
• A 50% reduction in mean time to resolve (MTTR)

Investor confidence: The rapid Series A — arriving six months after the seed round — with participation from DST Global Partners, Quantumlight, Picus, and Databricks Ventures alongside returning investors, suggests strong institutional conviction in the AI-era observability thesis.

The Bottom Line

Tsuga is one of the better-funded European startups in the observability space, with $45M in total funding, a team of experienced Datadog alumni, and a clear thesis: observable infrastructure must be rebuilt for AI agent workloads. Its BYOC deployment model, pricing strategy, and agent-native design represent a genuine architectural departure from the Datadog-Dynatrace duopoly. The early customer results — particularly the 30% reduction in MTTD and 50% reduction in MTTR at Le Monde — provide encouraging signal, though the company remains early in its lifecycle. For engineering teams evaluating whether to build with AI agents or invest in observability for AI workloads, Tsuga represents a credible alternative that is worth watching closely.