Boidwatch
Boidwatch is an autonomous UX-evaluation product that dispatches Census-derived, LLM-driven personas through live web pages in real Chromium sessions. It records what synthetic visitors tried, perceived, felt, remembered, and where they got stuck, then turns the traces into client-ready UX reports.
At a Glance
- UX and design studios
- Digital agencies
- Conversion-rate-optimization teams
- Product teams and web builders
- +2 more
AI Tools by Boidwatch
(1)
Boidwatch
LLM Persona UX Audit Tool
Discussions
No discussions yet
Be the first to start a discussion about Boidwatch
Latest News
Boidwatch publishes 'Best usability testing tools in 2026' and details its synthetic-browser evidence model
Boidwatch CLI documentation fixes the example API-key prefix from fk_ to fg_
Boidwatch CLI v1.0.0 is released as a public command-line client
Boidwatch publishes the CLI distribution, installer, and proprietary-use license
Products & Services
Runs a flock of Census-derived personas through a supplied live URL in real Chromium sessions and produces a public-link report with funnel, friction findings, segment breakdown, and key-moment screenshots. Accounts receive free credits on signup.
Programmatic interface for creating and monitoring exploration runs, retrieving results, managing personas and site profiles, and checking workspace billing. It is designed for terminal, CI, and agent-driven workflows.
Public pure-HTTPS command-line client for hosted or self-hosted Boidwatch. It supports authentication, run creation/status/results/list/cancel, billing, personas, site profiles, JSON output, a machine-readable command manifest, and stable exit codes.
Market Position
Boidwatch positions itself between traditional human usability research and screenshot-only AI heuristic reviewers: it uses synthetic LLM personas but executes real browser sessions and supplies traceable artifacts. Its stated differentiator is client-ready, evidence-backed behavioral reporting from many population-weighted personas in minutes, while acknowledging that it does not replace real customers, accessibility audits, or device/network testing. Adjacent alternatives discussed by Boidwatch include Microsoft Clarity, Hotjar, FullStory, PostHog, UserTesting, Userlytics, Maze, Lyssna, Loop11, Optimal Workshop, Lighthouse, axe DevTools, and LLM heuristic reviewers.
Leadership
Founders
Jesse Chow
Product engineer and ex-founder. Jesse Chow's public GitHub profile describes him as someone who ships products end-to-end; he is the visible maintainer/contributor associated with the Boidwatch CLI repository, whose two listed contributors are Jesse Chow and Claude. LinkedIn search results describe him as having built Boidwatch's autonomous UX-testing engine and as a sole founder/creator.
Executive Team
Jesse Chow
Founder / product engineer
Public GitHub profile identifies Jesse Chow as a product engineer and ex-founder who ships products end-to-end. He is the named human contributor on the public Boidwatch CLI repository.
Founding Story
Boidwatch was created to make specialist UX audits faster and more evidence-based for studios, agencies, and CRO teams. Its initial vision is to run many Census-weighted synthetic personas through a live page before a client kickoff, capturing executable browser evidence rather than relying on invented personas, heatmaps, or prose-only AI opinions.
Business Model
Revenue Model
Credit-based usage for persona sessions, with credits purchased through Stripe, plus a monthly Agency Founding subscription. The Terms state that one credit typically corresponds to one agent session; the exact run cost is shown before submission.
Pricing Tiers
100 persona sessions.
300 persona sessions.
750 persona sessions.
1,500 persona sessions per month, white-label reports, unlimited client share links, and a direct line to the founder.
30 free credits on signup, described as enough for six quick checks; no card required.
Target Markets
- UX and design studios
- Digital agencies
- Conversion-rate-optimization teams
- Product teams and web builders
- Developers and coding-agent users working in Cursor, Claude Code, Bolt, or Lovable
- Teams evaluating live pages without enough traffic for reliable replay
- Fast UX audits at the start of a client engagement
- Conversion-rate optimization and landing-page evaluation
- Testing live websites before a kickoff call or stakeholder review
- Finding friction and drop-off in signup, trial, pricing, and form flows
- Generating client-ready evidence for studios and agencies
- Running UX evaluation from a terminal, CI pipeline, or coding-agent harness