On-Call Health: A "Health Report" for SREs, Catching Burnout Before It Crushes You

2026-02-12 | Product Hunt | Official Site | GitHub

The main interface is a burnout analysis dashboard. Dark navigation on the left, with four core modules on the right: Burnout Timeline, Health Trends, Team Risk Factor Radar, and Individual Risk Scores. Color-coded from green to red, it shows at a glance who is reaching their limit.

30-Second Judgment

What it does: It extracts signals from tools like PagerDuty, GitHub, Slack, and Jira, combines them with engineer self-assessments, and calculates a "Burnout Risk Score." Essentially, it’s a health check for on-call engineers, flagging potential breakdowns before they happen.

Is it worth it?: If you manage a team with an on-call rotation—absolutely. It’s currently the only tool based on a validated academic framework (Copenhagen Burnout Inventory), it's open-source, free, and self-hostable. While PagerDuty Analytics gives you raw numbers, this tells you who is struggling.

Three Key Questions

Is this for me?

• Target Audience: Engineering Managers, Tech Leads, VPs of Engineering in SRE/DevOps, and any engineer on a rotation.
• The Test: If you look at a PagerDuty schedule every week, or if you've lost talent due to on-call stress—this is for you.
• Use Cases:
  • A team member has had a heavy load for weeks -> Use this to catch the risk early.
  • Quarterly reviews -> Use this to quantify the impact of on-call on team health.
  • Onboarding new hires to the rotation -> Monitor their load to ensure they aren't overwhelmed.
  • Solo developer with no on-call? -> You don't need this.

Is it useful?

ROI Judgment: If your team is 5+ people on rotation, spending 30 minutes to deploy this is a steal. The cost of recruiting and training a replacement for a senior SRE who quits due to burnout is astronomical; this tool prevents that for free.

What's the "Wow" factor?

The Highlights:

• Personal Baseline Tracking: It doesn't compare you to others; it compares you to your own history. A veteran's capacity differs from a junior's—this design is brilliant.
• Multi-Signal Fusion: It looks beyond just the number of pages. It tracks after-hours activity, PR volume, code review frequency, and Slack activity patterns.
• AI Summaries: Automatically tells you "what changed and why," basically writing your weekly health report for you.

Real User Feedback:

"Anyone that's worked at a very fast paced company knows how real pager fatigue and burnout is. Especially with AI coding assistants accelerating the number of (complex) incidents, measuring this has never been more important." -- Product Hunt User

"SRE burnout rarely looks dramatic from the outside. It just looks like someone quietly disengaging. Tools that make that visible are long overdue." -- Product Hunt User

"I used to tell myself I was 'fine' on call. It was the slow erosion. The Slack pings during dinner. Not taking my prescribed sleeping pills 'just in case.' Planning weekends around a laptop." -- Product Hunt User

For Developers

Tech Stack

• Frontend: Web App (Hosted at oncallhealth.ai)
• Backend: Docker Compose deployment, supports self-hosting
• AI/Models: AI analysis for trends and drivers (Models not disclosed, but Rootly AI Labs is backed by Anthropic and Google DeepMind expertise)
• Auth: Google OAuth / GitHub OAuth
• Research Foundation: Copenhagen Burnout Inventory (CBI) + Maslach Burnout Inventory (MBI)

Core Implementation

The data layer connects to PagerDuty/Rootly for incident data (MTTR, severity, after-hours pages), GitHub for code activity (PRs, commits, reviews), Slack for communication patterns and check-ins, and Jira/Linear for workload (issue allocation). All signals aggregate into an OCH Score (0-100), compared against historical baselines rather than fixed thresholds. Tiers: 0-24 Balanced, 25-49 Risk, 50-74 Intervention, 75-100 Action.

Left: Bar chart showing CBI scores for each member. Middle: Radar chart analyzing team risk across five dimensions (intensity, response pressure, etc.). Right: Time-series graph tracking health trends.

Open Source Status

• Open Source: Fully open, GitHub repo Rootly-AI-Labs/On-Call-Health
• License: Forkable and customizable
• Alternatives: No direct open-source competitors. Grafana OnCall handles scheduling but not burnout detection, and its OSS version is entering maintenance mode (archiving March 2026).
• Build Difficulty: Medium-High. Integration is easy (standard APIs), but the burnout scoring model calibration is hard. Forking this is the fastest way to get started. Estimated 1-2 person-months to build from scratch.

Business Model

• Monetization: On-Call Health itself is not monetized; it is completely free.
• The Strategy: It acts as a lead magnet for Rootly’s paid incident management platform. Once you see the burnout issues via the free tool, you're more likely to consider Rootly’s full suite.
• Customer Base: Rootly serves NVIDIA, LinkedIn, Dropbox, Tripadvisor, etc.

Big Player Risk

PagerDuty has an Analytics module, but it's generic reporting. Opsgenie (Atlassian) is closing in 2027. incident.io lacks this specific feature. It's unlikely giants will build a standalone burnout product soon as it's often viewed as a "nice to have." Being open-source protects the tool's longevity.

For Product Managers

Pain Point Analysis

• The Problem: SRE burnout is a "boiling frog" issue—not one big incident, but daily micro-stress. 70% of SREs report burnout leading to turnover.
• Severity: High frequency + High demand. PagerDuty 2024 reports show a 16% increase in incidents. AI assistants speed up code production, which in turn speeds up production incidents. Tired engineers make slower decisions and quit more often.

User Persona

• Primary: Engineering Manager / VP of Engineering—needs data to justify decisions, not just "vibes."
• Secondary: The SREs themselves—they participate via self-reporting and can see their own trends.
• Scenario: Pulling up the On-Call Health dashboard during weekly incident reviews or quarterly planning to argue for more headcount.

Feature Breakdown

Competitive Differentiation

Key Takeaways

1. Apple Health-style Reporting: Using short, low-friction check-ins (like Apple's "State of Mind") significantly increases user participation.
2. Baseline vs. Threshold: Avoid "X pages = burnout." Tracking individual deviation from the norm is a much more accurate health metric.
3. Open-Source PLG: Using a free tool to build brand authority before converting to a paid platform is a masterclass in B2B SaaS growth.

For Tech Bloggers

Founder Story

• Sylvain Kalache: Lead at Rootly AI Labs. Former LinkedIn Senior SRE (co-patented self-healing infra), then co-founded Holberton School (backed by Jerry Yang and Solomon Hykes). His path from SRE to education and back to tooling is a great story.
• Spencer Cheng: Software Engineer at Rivian, contributing as a Fellow.
• Backing: Support from Anthropic and Google DeepMind experts gives this "side project" serious weight.

Discussion Angles

• "Can burnout be quantified?": OCH Score measures workload, not biology. Is it right to turn feelings into numbers? (The project notes it's not a medical tool).
• "Caring or Spying?": Is this a manager's tool for support or a new KPI for surveillance? The boundary is thin.
• "The AI Loop": AI speeds up code -> AI causes incidents -> AI detects burnout. There's a certain irony there worth exploring.

Hype Metrics

• PH Ranking: 107 votes.
• Hacker News: Gained traction via Show HN.
• Media: Covered by VMBlog and other tech outlets.
• Trend: Just released (Feb 11, 2026), currently on the rise.

For Early Adopters

Pricing Analysis

No paid version, no hidden costs. The only cost is your server and setup time.

Quick Start Guide

• Time: 30 mins (5 mins for hosted).
• Curve: Low.
• Steps:
  1. Visit oncallhealth.ai to try the mock data.
  2. To use real data: Fork the GitHub repo, set up .env (OAuth tokens).
  3. Run docker compose up -d.
  4. Connect PagerDuty/Rootly + GitHub + Slack.
  5. Wait for data collection and scoring.

The Catch

1. Beta Status: It's early; expect bugs.
2. OAuth Setup: Self-hosting requires Google/GitHub OAuth config, which might be tricky for beginners.
3. Not Diagnostic: It tracks workload trends, not clinical mental health.

Security & Privacy

• Storage: Self-hosted data stays on your hardware.
• Privacy: Requires team consensus as it tracks work patterns.
• Audit: Code is open for your own review.

For Investors

Market Analysis

• Market Size: Incident response was $2.92B in 2022; DevOps is hitting $86B by 2034.
• Growth: 500%+ growth in the DevOps sector (2024-2034).
• Drivers: AI code assistants = more code = more incidents = higher on-call pressure = higher demand for health tools.

Competitive Landscape

• Top Tier: PagerDuty, Datadog (General monitoring, no burnout focus).
• Mid Tier: incident.io, FireHydrant, Rootly (Management/Scheduling).
• New Entry: On-Call Health (Specialized burnout detection).
• Exiting: Opsgenie (Closing 2027).

Timing

• Why now?: AI is accelerating the incident cycle, mental health is a corporate priority, and SRE talent is at an all-time premium.
• Maturity: APIs are ready, academic frameworks (CBI) are proven, and AI is finally capable of meaningful automated analysis.

Team & Funding

• Rootly Team: Founded 2020, YC Alumni.
• Funding: $15.3M total. Series A ($12M) led by Renegade Partners in 2023.
• Investors: Y Combinator, Google Gradient Ventures, 8VC, etc.
• Customers: NVIDIA, LinkedIn, Dropbox, Replit.

Conclusion

On-Call Health does what everyone knew needed doing but no one actually did: it quantifies on-call burnout risk. It’s not a revolutionary new technology, but a smart combination of existing data (PD, GitHub, Slack), academic research (CBI), and AI analysis. Its open-source nature makes it an easy win for any engineering team.

Resource Links

2026-02-12 | Trend-Tracker v7.3