Composal GTM brain

category

market profile

• text: Faster code generation increases pressure on teams to validate actual user journeys before release.
• text: AI testing products increasingly explore browsers, maintain tests and report reproducible failures.

personas

• title: Engineering manager
  pains:

  • Release confidence depends on slow manual checks across critical flows.
  • A passing deployment can still break a customer's journey.
• title: QA lead
  pains:

  • Edge cases across roles and permissions are hard to cover consistently.
  • Reproducing an intermittent failure consumes debugging time.
• title: Developer using coding agents
  pains:

  • Generated changes need independent validation against intended behavior.
  • Bug reports without steps and logs slow down fixes.

icp

• text: Product engineering teams shipping web applications frequently, especially those using coding agents.
• text: SaaS or commerce teams with important multi-step flows, roles and permissions that need regression coverage.

message pillars

• text: Test from the user's perspective before release.
• text: Explore edge cases in parallel.
• text: Deliver evidence engineers can act on.

objections

• objection: How do we know a reported issue is real?
  response: Inspect recordings, logs and reproduction steps, and pilot against a known flow before making it a release gate.
• objection: Will the agents understand our product's intent?
  response: The product maps the application using a URL and connected product sources; validate coverage with your team.
• objection: We already have automated tests.
  response: Evaluate autonomous exploration of user journeys and edge cases as a complement to existing checks.

competitors

• name: Momentic
  why: AI QA platform that runs and maintains tests and reports bugs.
• name: TestSprite
  why: Agentic end-to-end testing after code changes.
• name: KaneAI
  why: AI testing agent for planning and maintaining end-to-end tests.

influencers

• name: Pandelis Zembashis
  url: https://www.ycombinator.com/founders/pandelis-zembashis
  why: CEO with prior experience building coding agents.

keywords

• AI QA agent
• autonomous browser testing
• AI end to end testing
• test coding agent output
• automated user journey testing
• regression testing agents

voice

• Direct and technical
• Outcome focused
• Practical and evidence oriented

signals

• text: A team is adopting coding agents and struggling to validate the faster stream of changes.
• text: Releases require manual regression testing of checkout, account or permission flows.

stack

• GitHub
• Notion
• PostHog
• Claude Code
• MCP

pricing

motion

channels

• developer documentation
• self-service
• sales
• YC directory

confidence

Founders

• name: Pandelis Zembashis
  title: Founder/CEO
  role: CEO
  bio: I'm co-founder and CEO of Composal. We help engineering teams continually test and validate what they build: our agents use the product, check real user flows and discover bugs before release. Previously, as CTO at Cosine, I built AI coding agents and the infrastructure behind them, including one of the first CLI harnesses, sandboxed execution from scratch and large-scale repository mirroring for data generation and model training. Now I help teams validate software at scale.
• name: Jonathan Davies
  title: Founder/CTO
  role: CTO
  bio: Co-Founder and CTO of Composal. Previously Senior Software Engineer at PapayaDash. Before that built software for Telefonica, Mitie, and Vodafone
• name: Jan Moehlmann
  title: Founder
  role: Founder
  bio: Founder at composal.ai (YC F26). Previously built commercial functions at 11x, Encord, Tractable