Deeptrace

What do they actually do

Deeptrace runs an AI "SRE" that watches production alerts (Slack‑first), automatically investigates across your logs, metrics, and code, and posts a concise root‑cause summary with evidence back into the Slack thread. Teams connect Slack and data sources during onboarding; the agent then auto‑triggers on alerts and returns likely causes in minutes, with links to the underlying data for quick verification deeptrace.com. The site lists several startup customers (e.g., Parafin, Rain, Fragment) deeptrace.com.

In practice today it focuses on triage and root‑cause summarization inside Slack; it does not yet replace engineers for complex remediation. The team’s roadmap is to extend from triage to resolution (e.g., suggested fixes/PRs) and broaden integrations, a direction they describe on their YC page and in recent coverage Y Combinator Forbes.

Who are their target customer(s)

• On‑call engineer (SRE or backend) at a startup: Frequent context switching across tools to pull logs/metrics and find likely causes; minutes lost during incidents that delay fixes. Deeptrace aims to auto‑investigate and post short, evidence‑backed summaries in Slack deeptrace.com.
• Engineering lead / CTO at a small or growing company: Team productivity is drained by firefighting and they can’t add headcount; they want faster triage and fewer interruptions so developers can keep shipping deeptrace.com.
• SRE or ops engineer at a small‑to‑mid company with limited headcount: Alert fatigue and repeated incidents with scattered historical context; needs tooling that surfaces the right logs and past signals quickly to avoid redoing investigations deeptrace.com Y Combinator.
• Platform/DevOps owner responsible for observability: Must connect logs, traces, and repos to automation while managing permissions, data exposure, and compliance. Wants broad integrations and clear security controls deeptrace.com.
• Incident commander / rotation owner: Needs concise, evidence‑backed summaries that enable quick handoffs so another engineer can pick up an incident fast. Prefers Slack‑first context with direct links to relevant data deeptrace.com.

How would they acquire their first 10, 50, and 100 customers

• First 10: Target YC alumni and similar startups for free, white‑glove pilots; integrate Slack + core data sources and deliver live root‑cause summaries in their workflows to generate early quotes and case studies deeptrace.com Y Combinator.
• First 50: Ship a Slack app listing and self‑serve trial with templates for common observability stacks; run webinars, sponsor SRE/DevOps communities, publish 2–3 customer stories, and start lightweight integrations/partnerships with popular monitoring/logs tools deeptrace.com.
• First 100: Convert pilots into team plans with security docs and an onboarding checklist for adding repos/log sources; pair with targeted outbound to engineering leads using case studies and a pilot SLA to de‑risk outcomes deeptrace.com Y Combinator.

What is the rough total addressable market

Top-down context:

On‑call automation sits inside overlapping observability, incident‑management, and AIOps spend. Public estimates put AIOps at ~USD 5.3B in 2024, observability around USD 2.9B in 2025, and incident‑management software in the low‑to‑mid billions; de‑duplicating overlap yields a realistic USD 4–7B pool relevant to automating triage/remediation GMI Mordor Intelligence VMR ResearchAndMarkets. PagerDuty’s ~$467M FY2025 revenue shows the incident‑oriented platform ceiling and enterprise weighting PagerDuty IR.

Bottom-up calculation:

Assume 50k–100k SMB/startup engineering teams with on‑call workflows that already pay for observability/alerting, and an annual budget of ~$10k–$15k for automation that reduces triage time. That implies an SMB/startup SAM of roughly $0.5–1.5B, consistent with the top‑down view.

Assumptions:

• Significant buyer overlap across observability, incident management, and AIOps (roughly 40–60%), so simple summation would double‑count.
• SMBs/scale‑ups represent ~10–30% of the combined relevant software spend for this use case today.
• Typical ACV for on‑call automation in SMBs is ~$10k–$15k when positioned alongside existing observability/incident tools.

Who are some of their notable competitors

• incident.io: Slack‑first incident manager now offering an “AI SRE” that triages alerts, pulls telemetry and past context, and suggests next steps inside Slack AI SRE Slack integration.
• Rootly: Incident platform with deep Slack workflows and AI features for auto summaries, investigations, and mitigation/resolution suggestions AI SRE Slack docs.
• FireHydrant: Incident orchestration tool with AI‑generated summaries, suggested updates, and retro drafts to reduce manual coordination and documentation AI overview Docs.
• BigPanda: AIOps/event correlation for larger ops teams; automatically groups/triages alerts and surfaces likely root causes with suggested actions Advanced Insight AI detection/response.
• PagerDuty: Broad on‑call/alerting platform with growing gen‑AI and AIOps features; integrates with Slack and can automate parts of incident workflows Generative AI Slack integration.