Prime Radiant
Prime Radiant is an applied AI research lab founded by Jesse Vincent in 2025, studying how production software gets built when AI coding agents do the implementation and how to manage those agents safely at scale.
About
Prime Radiant is an applied AI research lab founded by Jesse Vincent in 2025, studying how production software gets built when AI coding agents do the implementation and how to manage those agents safely at scale. It created and maintains Superpowers, an open-source agentic skills framework (and the SKILL.md format) for coding agents such as Claude Code, with over 500,000 installs.
Comparable companies· by market
This company
BusinessLatest roundTotal raisedDate- Prime Radiant
Prime Radiant is an applied AI research lab founded by Jesse Vincent in 2025, studying how production software gets built when AI coding agents do the implementation and how to manage those agents safely at scale. It created and maintains Superpowers, an open-source agentic skills framework (and the SKILL.md format) for coding agents such as Claude Code, with over 500,000 installs.
—
United States
BusinessLatest roundTotal raisedDate- SylphAI
AI coding/agent engineering platform AdaL (CLI, agentic IDE, coding and research worker agents, browser automation, Slack/GitHub integrations, skills marketplace), built on its in-house open-source LLM pipeline auto-optimization framework AdalFlow; customers are developers and engineering teams.
$2M
$2M - BMad Method
Open-source spec-driven framework for AI agent software development, with paid training, certification and enterprise consulting, helping engineering teams adopt AI-native development with governance and traceability and tackling the context drift and runaway requirements of "vibe coding".
— - exe.dev
"Real computers" for developers and AI coding agents: per-second-billed disposable sandbox VMs (Sandbox), persistent SSH-accessible Linux VPS, cloud Devbox and the web-based AI coding agent Shelley; API and ssh exe.dev access, on machines in its own data centers. Customers include Antithesis, which uses it as a sandbox for agentic software testing.
$35M
$35M
China
BusinessLatest roundTotal raisedDate- Verdent
AI-native software company focused on "Agentic Coding" (coding agent) tools; product: Verdent (VS Code extension)
Series A · $30M · 2025-11
$30M - aiXcoder
AI virtual programming robot
Series A · $7M · 2023-09
$7M - 一平方科技
AI coding solutions and R&D in core NLP and large model technologies; its product HiBug provides efficient AI coding tools
Series A · 2024-10
—
Europe
BusinessLatest roundTotal raisedDate- GitHits
Version-aware indexing of open-source code/packages and code example retrieval for AI coding agents (CLI + MCP integration)
— - VirtusLab
Custom software engineering and consulting (reactive systems, data engineering, cloud native, frontend), Scala open-source ecosystem maintenance, AI-assisted development and agent infrastructure tools
— - Namespace
Build and test compute infrastructure for developers and AI coding agents
Series B · $65M · 2026-10
$65M
Other / unspecified
BusinessLatest roundTotal raisedDate- ClackyAI
AI coding cloud IDE Clacky.ai and open-source agent framework OpenClacky, automating requirements-to-code/PR for developers, tech teams and non-technical founders
— - gjalla
Architecture governance and observability SaaS for AI coding agents: doc drift, policy checks and audit trails, token spend attribution
— - AgentGrid
Desktop workbench for developers to orchestrate multiple AI coding agents, running and coordinating Claude Code, Codex, Cursor and others in parallel on an infinite canvas
—
Confirmed rounds only. Matched on subsector, sector, stage and funding scale — each verifiable on the company profiles.
Investors that may fit
Andreessen Horowitz
15 Developer Tools deals in the last 24 months · 15 in Silicon Valley
Request intro →General Catalyst
12 Developer Tools deals in the last 24 months · 10 in Silicon Valley
Request intro →Ranked by each firm’s deals in this sector over the last 24 months, weighted by nearby stage and shared geography. Indicative only — no rating judgment.