
What they do:
The IDE for Tokenmaxxers
Quick facts:
HQ: Boston, MA, United States
This list has 493 startups and was last updated on Oct 15, 2026. 123 of them are hiring right now. Biggest hubs: San Francisco (302), New York (44), and London (16). Largest teams: Fivetran, PagerDuty, and Firebase. Latest round: Salience Labs raised $125M (Oct 14, 2026).
Survival statistics →452 open roles →

What they do:
The IDE for Tokenmaxxers
Quick facts:
HQ: Boston, MA, United States

What they do:
Co-designing the fastest intelligence
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Effortlessly deploy cloud software factories.
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Self-optimizing neocloud that cuts LLM costs by 100x
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Market for future inference capacity
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Derivatives Exchange for Compute
Quick facts:
HQ: New York, NY, United States

What they do:
Expert computer-use data for frontier AI labs
Quick facts:

What they do:
We find where AI agents fail, then train better models to run them.
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Training company-specific models that approve AI actions.
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Autonomous rovers to make the ‘Google Maps’ of underground utilities
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Open source AI gateway that turns your traffic into better models
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Intelligent Cloud for AI workloads
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Building in-house, frontier AI labs for companies
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Fast inference for voice
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Turn any website into an MCP for your agents, in minutes
Quick facts:
HQ: San Francisco, CA, United States

What they do:
reducing agent reasoning into a graph search problem
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Agent swarms that make long-running tasks exponentially faster
Quick facts:
HQ: London, United Kingdom

What they do:
Durable state for AI agents
Quick facts:
HQ: New York, NY, United States

What they do:
Verify that the expected person is in control before key actions.
Quick facts:

What they do:
Make your LLMs 3x faster.
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Turning messy inboxes into ERP-Ready Data
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Verify AI-generated code before you ship it
Quick facts:
HQ: San Francisco, CA, United States

What they do:
AI for public records request fulfillment
Quick facts:
HQ: Seattle, WA, United States

What they do:
Let your camera understand the world
Quick facts:
HQ: Denver, CO, United States

What they do:
Agentic AI Security and Governance
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Managed GPU-accelerated Spark + SQL at 2x speed and half the cost
Quick facts:
HQ: New York, NY, United States

What they do:
Infrastructure Cartography
Quick facts:
HQ: Boston, MA, United States

What they do:
AI Data Analyst for fast-growing teams
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Ship software faster with portable, self contained virtual machines.
Quick facts:
HQ: San Francisco, CA, United States

What they do:
The Git forge for the next hundred billion commits
Quick facts:
HQ: San Francisco, CA, United States

What they do:
The AI gateway for cheaper, faster, permissioned agents.
Quick facts:
HQ: San Francisco, CA, United States

What they do:
AWS for Agents
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Managed Agents API
Quick facts:
HQ: San Francisco, CA, United States

What they do:
The open source tool gateway
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Monetize your agent's traces.
Quick facts:
HQ: San Francisco, CA, United States

What they do:
Open-model inference provider for Europe, private by default.
Quick facts: