In 2021, ThermoMapIQ co-founders Marcus Klein and Priya Yuen were both on call at a hyperscale colocation facility when a thermal incident took down 14 compute nodes in a single rack row. The root cause was a hot-aisle containment bypass that had gone undetected for weeks — invisible to every monitoring tool in the stack.
Three dashboards, two ticketing systems, one aging CMDB, and none of them could draw a straight line between a misconfigured blanking panel, a heat signature, and a throttling server. The data existed. The spatial context did not.
They left their jobs six months later. The first version of ThermoMapIQ was running on a converted 1U rack in Marcus's garage by early 2022, ingesting SNMP feeds and plotting device temperature readings on hand-drawn floor plan exports. By the end of that year, three enterprise customers were in pilot. The spatial layer they were building turned out to be the piece that everyone needed but nobody had built.
Give every IT operations team the spatial intelligence layer their infrastructure has always lacked — so that thermal incidents, ghost assets, and placement inefficiencies are solved before they become outages.
A world where no data center operator makes a rack placement, cooling, or capacity decision without a live, AI-enriched spatial model of their entire physical estate — from a single edge closet to a 50,000 sq ft multi-tenant facility.
These are not wall posters. They are the criteria we apply when designing features, resolving trade-offs, and deciding which customer asks to prioritize.
Cached or manually-updated data is worthless during a thermal incident. Every data point in ThermoMapIQ is live — updated on the second, not the hour. We engineered our ingestion pipeline to handle over 10 million sensor reads per day without degrading UI responsiveness.
Infrastructure intelligence should connect to any sensor, protocol, or cloud. Our REST and MQTT APIs are open-spec, all data is exportable, and Enterprise customers can run the full stack on-premises with zero data leaving their facility.
Alerting after a server throttles is too late. Our AI models flag thermal risk 20 minutes before it materializes. We obsess over lead time because that gap is the difference between a prevented outage and a missed SLA. Prediction is not a feature — it is our core premise.
A timeline of the moments that defined how we built ThermoMapIQ into the platform it is today.
A hot-aisle bypass goes undetected for weeks and takes down 14 compute nodes. Marcus and Priya spend three days doing a post-mortem and realize no tool in the stack had spatial context. They start sketching the first version of ThermoMapIQ on a whiteboard in the colocation operations room.
The first working prototype ingests SNMP feeds and overlays device temperatures on SVG floor plan exports. Three enterprise customers begin closed pilots. A $2.4M seed round closes in November, led by Basecamp Ventures, and the company incorporates as ThermoMapIQ, Inc. in San Francisco.
The first AI-powered thermal prediction model ships in March — trained on 18 months of sensor data from pilot customers. The platform launches to general availability in Q2. By year-end, ThermoMapIQ has 40 paying customers, 60,000 assets under management, and an NPS of 72.
A $14M Series A closes in Q1, led by Accel Partners. The Enterprise on-premises edition ships in Q3, enabling air-gapped deployments for government and defense customers. Customer count passes 200. The team grows from 8 to 45 across San Francisco, London, and Singapore.
Multi-site federation ships, allowing global enterprises to manage hundreds of facilities from a single pane of glass. The platform surpasses 500 enterprise customers and 3 million assets under management. ThermoMapIQ is listed in the Gartner Emerging Technologies Hype Cycle for Infrastructure Monitoring.
Thermal AI v3 ships with predictive cooling recommendations and PUE optimization suggestions. The team reaches 60+ engineers globally. We are pursuing SOC 2 Type II + ISO 27001 certification and are expanding API coverage to include Redfish v1.14 and OpenBMC integrations — with carbon footprint reporting on the roadmap.
Our team of 60+ spans distributed systems, thermal physics, machine learning, and enterprise IT operations — based in San Francisco, London, and Singapore.
15 years in hyperscale data center operations. Former Director of Infrastructure at Fastly. Holds patents in real-time network topology visualization.
PhD in Distributed Systems, Stanford. Former principal engineer at Cloudflare. Led the design of our real-time sensor ingestion pipeline handling 10M+ reads/day.
15 years building observability platforms at Datadog and New Relic. Architected ThermoMapIQ's multi-tenant infrastructure and enterprise on-premises deployment stack.
Former research scientist at DeepMind, specializing in physics-informed neural networks. Leads the thermal prediction model and placement optimization research teams.
10 years in enterprise SaaS product management at PagerDuty and Splunk. Owns the platform roadmap and customer feedback loop that has driven 72+ NPS for three consecutive years.
Built and scaled enterprise sales teams at Zscaler and Rubrik. Oversees the 500+ enterprise customer portfolio and the strategic accounts program covering Fortune 500 deployments.
We built security and compliance into the platform architecture — not as afterthoughts bolted on post-launch.
Annual third-party audit covering security, availability, and confidentiality. Reports available to enterprise customers under NDA.
All customer data encrypted at rest using AES-256-GCM and in transit via TLS 1.3. Zero plaintext storage of sensitive credentials or API keys.
Full compliance with EU GDPR, California CCPA/CPRA, and Brazil LGPD. Data residency options available for EU and APAC customers.
Contractual uptime guarantee for Thermal Pro and Enterprise subscribers, with service credits for SLA violations. Status page at status.thermomapiq.com.
Enterprise customers can deploy ThermoMapIQ entirely within their own infrastructure. Air-gapped deployments supported for government and defense workloads.
Quarterly penetration tests by independent security firms. Bug bounty program managed through HackerOne. Responsible disclosure policy available at thermomapiq.com/security.
Ready to see ThermoMapIQ in action with your own infrastructure?
Request an Enterprise Demo