Velvet – Store OpenAI requests in your own DB
Hey HN! We’re Emma and Chris, founders of Velvet (https://www.usevelvet.com). Velvet proxies OpenAI calls and stores the requests and responses in your PostgreSQL database. That way, you can analyze logs with SQL (instead of a clunky UI). You can also set headers to add caching and metadata (for analysis). Backstory: We started by building some more general AI data tools (like a text-to-SQL editor). We were frustrated by the lack of basic LLM infrastructure, so ended up pivoting to focus on the tooling we wanted. So many existing apps, like Helicone, were hard to use as power…
Velvet – Store OpenAI requests in your own DB launched on September 24, 2024 with 109 votes, #492 of 5,281 launches that month and more than 90% of that year's launches. The site was still online in September 2026, 2 years on.
In plain words
Velvet is a proxy service that intercepts OpenAI API calls and stores requests and responses directly in a user's PostgreSQL database. This allows developers to query and analyze their LLM usage patterns using SQL rather than relying on third-party interfaces. The platform supports adding custom headers for caching and metadata enrichment, and is built on Cloudflare Workers to minimize latency overhead. It is designed for developers and teams who need direct database access to their AI infrastructure logs.
written from the facts on this page · September 2026
From the sources
Velvet is now part of Arize, the enterprise platform for AI evaluation and observability.
Velvet made a developer gateway to analyze, evaluate, and monitor AI-powered features. Arize, the enterprise platform for AI evaluation and observability, acquired Velvet in 2025. Emma and Chris were the founders of Velvet, and joined forces with Arize to accelerate adoption of Arize's unified AI platform. Unified Observability and Evaluation Platform for AI. Arize helps you accelerate development of AI apps and agents, then perfect them in production. Open-source LLM tracing and evaluation to accelerate AI development. Seamlessly evaluate, experiment, and optimize AI applications in real time. LLM Gateway to provide model access, fallbacks, and spend tracking across 100+ LLMs. All in the…from usevelvet.com
In the maker’s words, at launch
Hey HN! We’re Emma and Chris, founders of Velvet (https://www.usevelvet.com). Velvet proxies OpenAI calls and stores the requests and responses in your PostgreSQL database. That way, you can analyze logs with SQL (instead of a clunky UI). You can also set headers to add caching and metadata (for analysis).…
Does a similar job
all alternatives →
SnapQL – Desktop app to query Postgres with AI2025 · github.com · ▲104SnapQL is an open-source desktop app (built with Electron) that lets you query your Postgres database using natural language. It’s schema-aware, so you don’t need to copy-paste your schema or write complex SQL by hand. Everything runs locally — your OpenAI API key, your data, and your queries — so it's secure and private. Just connect your DB, describe what you want, and SnapQL writes and runs the SQL for you.
I open-sourced the in-memory PostgreSQL I built at work for E2E tests2024 · github.com · ▲379In-memory Postgres for unit/E2E tests. Contribute to hexclave/pgmock development by creating an account on GitHub.
Hasura – Fast, instant GraphQL APIs on Postgres2018 · github.com · ▲382Blazing fast, instant realtime GraphQL APIs on all your data with fine grained access control, also trigger webhooks on database events. - hasura/graphql-engine
Laminar – Open-Source DataDog + PostHog for LLM Apps, Built in Rust2024 · github.com · ▲203Hey HN, we’re Robert, Din and Temirlan from Laminar (https://www.lmnr.ai), an open-source observability and analytics platform for complex LLM apps. It’s designed to be fast, reliable, and scalable. The stack is RabbitMQ for message queues, Postgres for storage, Clickhouse for analytics, Qdrant for semantic search - all powered by Rust. How is Laminar different from the swarm of other “LLM observability” platforms? On the observability part, we’re focused on handling full execution traces, not just LLM calls. We built a Rust ingestor for OpenTelemetry (Otel) spans with GenAI…
More ai this month
the category →
Paintings by AI models: programs run through a simulation of oil paint.
AI · 22d ago · stillwet.art
FastRouter.ai▲364Route requests to the right LLM for cost, latency & quality
AI · 23d ago · fastrouter.ai
Source: Show HN post ↗ · launch text quoted in part
Launched alongside, September 2024
the whole month →
BeforeSunset AI 2.0▲1,267Personalized AI daily planning that suits your life
AI · 2024 · beforesunset.ai
