The Ferr blog
Launches, engineering notes, and hands-on guides for building browser agents, scrapers, and automations in the cloud.
Launch Week v3: everything we shipped
Five days, five launches. Here is the full recap of Launch Week v3, from faster cold starts to Ferr Skills.
A simpler, usage-based pricing update
We are moving to pricing that tracks browser hours, so what you pay follows what you actually run.
Meet Atlas: the Ferr session runtime
Atlas is the orchestration layer that turns a single browser call into a managed fleet.
Ferr SDKs for Rust and Go
Typed, idiomatic clients for two languages systems teams live in — both wrapping the Sessions API.
Dedicated IPs for stable, long-running jobs
Route traffic through addresses you control, per project, and keep long crawls stable.
Stealth mode: browsers that look human
Fingerprints, header hygiene, and human-like timing, on by default so your browsers blend in.
Introducing Projects: organize your browsers
Group sessions, keys, IPs, and profiles under a Project so teams can share infrastructure cleanly.
Building a durable research agent on Ferr
A walkthrough of an agent that researches for hours, survives restarts, and never loses its place.
Agent logs: see what your browser did
Structured network and console logs on every session, ready the moment something goes wrong.
Agent traces: replay every session
A scrubable timeline of everything a session did, with a link you can share with your team.
How modern anti-bot defenses actually work
Fingerprinting, behavioral signals, and challenge flows — a field guide to what blocks browsers today.
A beginner's guide to Ferr
From zero to your first cloud browser in a few minutes, with the code to copy.
We rebuilt the browser-agent leaderboards
A fairer, reproducible way to compare browser agents on real tasks.
An autopsy of a Claude Code deep-research run
We traced a long deep-research session step by step to see where the time and tokens went.
Wiring Ferr and Resend into a daily newsletter
A weekend build: an agent that browses the web with Ferr and emails a digest with Resend.
Dane's journey to Ferr
How one indie developer went from brittle local scripts to a fleet of cloud browsers.
Eight ways to build a browser agent
A tour of the patterns teams use to put a real browser behind an LLM.
Errors are the interface agents actually read
If your API errors are vague, your agent will be too. A case for error messages written for machines.
Give Claude Code a real browser
Add Ferr as a tool so your coding agent can open pages, click, and read the live web.
The happy path for automating the web with Ferr
The smallest set of steps that gets a reliable automation into production.
Headless browsers and cloud scraping, explained
What a headless browser is, why plain HTTP is not enough, and how to scale the whole thing.
Introducing Ferr Skills
Package a repeatable browser task as a named, versioned Skill your agents can call.
Launch Week v2: the recap
Everything that shipped during Launch Week v2, in one place.
Launch Week v2 begins
Five more days of launches. Here is the theme and what to expect from Launch Week v2.
Mobile mode: emulate phones at scale
Render pages as a phone would, with the right viewport, touch, and user agent.
More browsers, more regions
We expanded capacity and added regions so your sessions start closer to your targets.
How Notte runs its web agents on Ferr
A look at how the Notte team uses Ferr sessions to power their web agents.
Let your AI agent fill the forms with OpenClaw + Ferr
Combine the OpenClaw agent framework with a Ferr browser to automate real form flows.
Taking browser agents to production
The checklist we wish every team had before shipping a browser agent to real users.
Browser profiles: keep your agents logged in
Save cookies, storage, and auth state to a profile and reuse it across sessions.
Benchmarking remote browsers
How we measure session start time, throughput, and stability — and what the numbers say.
Self-hosting Ferr Browser
Run the open-source Ferr Browser on your own infrastructure, from a laptop to a cluster.
Ship browser automation to users with Ferr and Convex
Store state in Convex, run browsers with Ferr, and give your users automations that just work.
The Ferr CLI and agent skill
Drive Ferr from your terminal, and give your agent the same commands as a Skill.
Ferr is now a native browser provider in Hermes
Select Ferr as the browser backend in Hermes and get cloud sessions with no extra setup.
The Ferr Node SDK
A typed, promise-first client for driving Ferr from Node and the browser.
The Ferr Python SDK
The Pythonic way to launch and drive cloud browsers, with sync and async clients.
Ferr vs Browserbase: a practical comparison
How the two browser platforms differ on stealth, observability, openness, and pricing.
Ferr vs Kernel
Two takes on browser infrastructure for agents, compared on the things that matter in production.
Top 7 headless browser automations that actually work
Seven automations that reliably pay for themselves, with notes on how to build each.
We gave Pi Agent a real browser
What happened when we connected the Pi agent framework to a Ferr session.
WebRTC support comes to Ferr
Sessions now handle WebRTC, so audio, video, and real-time features work in a Ferr browser.
What is a CAPTCHA solver?
A plain-language explanation of CAPTCHA challenges and how automated flows handle them.
Why browser agents fail in production
The failure modes that only appear at scale, and the patterns that prevent them.
Why we built Agent Traces
The debugging story that pushed us to record every session as a replayable trace.
Why we built inactivity timeouts
A small feature that quietly saves you money and keeps stuck sessions from lingering.