How DevTools companies build with AI — which foundational models they run and how directly, ranked by proximity to the model. Avg Foundation Proximity Score 50/100.
DevTools · Multi-Cloud · Native (own / self-hosted weights)
Cursor runs a multi-cloud, event-driven architecture with most CPU/backend services on AWS, major GPU inference clusters on Azure plus newer GPU clouds, and specialized AI workloads on GCP, fronted by stateless gateways, queues, and GPU worker fleets for LLM inference and agents.[6][10] Workspace data and embeddings flow through turbopuffer, Pinecone, and Kafka-compatible streams, with Temporal-based orchestration for agent workflows and large multi-GPU clusters (including Fireworks AI-hosted proprietary fine-tuned models) serving code-focused LLMs.[4][6][7][8][10]
DevTools · AWS · Direct API
API platform; Postbot AI on OpenAI over API specs/collections.
DevTools · GCP · Direct API
Cloud IDE whose Agent is built primarily on Anthropic Claude.
DevTools · AWS · Direct API
Internal-tools builder with multi-model AI app components.
DevTools · GCP · Direct API
Error-monitoring; Autofix/Seer uses OpenAI + Anthropic over stack traces.
DevTools · AWS · Direct API
Frontend cloud built atop AWS, exposing a global edge + serverless/Functions layer; stewards the open-source Next.js framework. v0 AI is built on OpenAI/Anthropic models.
DevTools · AWS · Cloud-hosted (Bedrock/Vertex/Azure)
Jira/Confluence migrated to an AWS-native multi-tenant platform (Micros) on PostgreSQL + DynamoDB. Atlassian Intelligence wraps OpenAI behind a tenancy/governance layer.
DevTools · Azure · Cloud-hosted (Bedrock/Vertex/Azure)
GitHub runs on a large-scale, multi-service production platform hosted on Microsoft Azure, with operational patterns documented by GitHub/Microsoft materials and independent technical writeups. The data layer is commonly described as MySQL at scale with Vitess plus Redis and Elasticsearch, while GitHub also publishes substantial open-source infrastructure and developer tooling.
DevTools · AWS · Middleware / wrapper
Postgres backend-as-a-service; AI via pgvector + partner embeddings.
DevTools · GCP · Off the LLM map (in-house non-LLM ML)
GitLab.com is primarily hosted on Google Cloud Platform in the us-east1 region, with most application services running on multiple Kubernetes clusters and core datastores (PostgreSQL, Gitaly, Redis, Elasticsearch) running outside Kubernetes. Secret management uses Google KMS for GCP services and Chef Encrypted Data Bags for other host secrets.
DevTools · AWS · Off the LLM map (in-house non-LLM ML)
Now part of IBM; HashiCorp Cloud Platform (HCP) for Terraform/Vault/Consul runs on AWS. Core tools are widely-adopted open-source (BSL-licensed) infra-as-code and secrets management.
DevTools · GCP · Off the LLM map (in-house non-LLM ML)
Linear’s CTO has described the production stack as running on Google Cloud with Kubernetes, using React and Node/TypeScript, with Postgres for primary storage, Redis for queuing, and MongoDB for caching scenarios. The only clearly documented AI-related capability in the provided sources is product-level “AI workflows”; no source here establishes a specific foundational LLM provider in the core production architecture, so model_links is empty.