Halmos Labs GTM brain

category

market profile

• text: Computational drug discovery teams must turn agent-generated hypotheses into reproducible physical experiments.
• text: Laboratory sourcing and contracting create friction between an in-silico result and wet-lab feedback.

personas

• title: Head of computational biology
  pains:

  • Promising model outputs wait for a suitable lab and experiment design
  • Results need a traceable link back to computational hypotheses
• title: Founder of an AI drug discovery startup
  pains:

  • Limited in-house wet-lab capacity
  • Coordination and legal work distract from testing candidates
• title: Research automation engineer
  pains:

  • Agents cannot independently route requests and retrieve experimental data
  • Human approval must fit into automated workflows

icp

• text: AI-native biotech startups and computational drug discovery teams with external wet-lab needs; account size is unconfirmed.
• text: Research groups developing agent workflows with Claude, ChatGPT, Codex or Cursor and needing physical experiments.

message pillars

• text: Close the loop between agents and real-world experiments.
• text: Reduce lab sourcing and contracting overhead while retaining human approval.
• text: Make agent reasoning inspectable and self-checking.

objections

• objection: Can an AI-designed experiment be trusted?
  response: Start with a bounded experiment, review the request and inspect assumptions and results; outcomes are not proof of a marketed drug.
• objection: Will our team lose control over lab orders?
  response: The published workflow includes a person reviewing and sending each request.
• objection: Can it reach a lab with the right capability?
  response: Ask for a sourced lab match and verify coverage for the specific assay before committing.

competitors

• name: Emerald Cloud Lab
  website: https://www.emeraldcloudlab.com/
  why: Remote automated life science experiments provide an alternative route to wet-lab validation.
• name: Strateos
  website: https://strateos.com/
  why: Automated labs support drug discovery workflows and experimental execution.
• name: Superdrugs
  website: https://superdrugs.life/
  why: Agents design discovery plans and route candidates toward synthesis.

influencers

• name: Mattias Akke
  url: https://www.ycombinator.com/founders/mattias-akke
  why: Halmos founder with drug discovery machine learning research experience.

keywords

• AI drug discovery agents
• cloud lab experiments
• wet lab validation AI
• automated drug discovery
• outsource biotech experiments
• physical validation MCP

voice

• scientific and reflective
• ambitious
• transparent about validation

signals

• text: A team is hiring computational biologists or building agents that generate testable drug candidates.
• text: Researchers need an external assay or lab and are using MCP-compatible agent tools.

stack

• MCP
• Claude
• ChatGPT
• Codex
• Cursor

pricing

motion

channels

• website
• research publications

confidence

Founders

• name: Mattias Akke
  title: Founder/CEO
  role: CEO
  bio: Founder at Halmos Labs. Previously ML research intern at MIT (Gómez-Bombarelli's lab) and AstraZeneca, developing frontier models for drug discovery. Published at NeurIPS on Bayesian optimization with LLMs. Research intern at Cambridge in Tuomas Knowles' group studying protein misfolding diseases. Chaired Kodsport Sverige, Sweden's largest competitive programming organization.
• name: Linus von Ekensteen Löfgren
  title: Founder/CTO
  role: CTO
  bio: Founder at Halmos Labs. Previously Desk Quant at Danske Bank, building pricing algorithms and risk models across all asset classes. Conducted LLM security research at AI Sweden and quantum computing research at Chalmers. MS Engineering Physics and BA Economics from Lund University.