Key takeaways
- 27K GitHub stars in ~14 months — among the fastest-growing open-source agent builders
- YC X25 batch with a $7M Series A led by Standard Capital, backed by Paul Graham and Perplexity
- Figma-like visual canvas differentiates from code-first frameworks like LangChain/CrewAI
- 1,000+ integrations and multi-LLM support position it as an agent-agnostic orchestration layer
FAQ
What is Sim.ai?
An open-source platform (Apache 2.0) for building AI agent workflows via a visual drag-and-drop canvas. Think Figma for agent pipelines — connect LLMs, tools, and data sources without writing boilerplate.
How does Sim.ai compare to n8n or Langflow?
Sim is purpose-built for agentic AI workflows with deep LLM integration, structured output, and granular tool-use control. n8n is broader workflow automation; Langflow is LangChain-specific. Sim sits in between — visual but AI-native.
Who founded Sim.ai and when?
Co-founded in 2025 by Emir Karabeg (CEO) and Waleed Latif (CTO), both UC Berkeley alumni. Emir previously built WorkNinja (300K+ MAUs). They graduated from YC's X25 batch.
What's Sim.ai's funding status?
$7M Series A (November 2025) led by Standard Capital, with participation from Paul Graham, Perplexity, SV Angel, Y Combinator, Ali Rowghani, Liquid 2 Ventures, and others.
Executive Summary
Sim.ai (formerly Sim Studio) is a YC-backed open-source platform for building, deploying, and orchestrating AI agent workflows through a visual canvas interface. Founded in 2025 by two UC Berkeley alumni, the company has grown to 27K GitHub stars and claims 100K+ builders on its platform. With $7M in Series A funding from an impressive roster (Paul Graham, Perplexity, SV Angel), Sim occupies a unique position: more AI-native than general workflow tools like n8n, more visual than code-first frameworks like LangChain.
| Attribute | Value |
|---|---|
| Company | Sim Studio, Inc. |
| Domain | sim.ai |
| Founded | 2025 |
| HQ | San Francisco, CA |
| Team Size | ~5 employees |
| Funding | $7M Series A (Nov 2025) |
| Lead Investor | Standard Capital |
| YC Batch | X25 |
| GitHub Stars | 27.1K |
| License | Apache 2.0 |
| Claimed Users | 100,000+ builders |
Founders
Emir Karabeg — Co-Founder & CEO
- UC Berkeley — Data Science & Cognitive Science
- NLP research (machine translation) at Berkeley
- Previously built WorkNinja, an AI learning platform that scaled to 300K+ MAUs
Waleed Latif — Co-Founder & CTO
- UC Berkeley
- Technical co-founder with engineering background
Both founders are relatively young/early-career. The WorkNinja track record (300K MAUs) is notable for a pre-YC venture — it shows Emir can build products people actually use.
Product Overview
Sim's core product is a visual agent workflow builder — a drag-and-drop canvas (compared to Figma's interface paradigm) where users connect LLM nodes, tool integrations, and data sources into executable agent pipelines.
Key Capabilities
| Capability | Description |
|---|---|
| Visual Canvas | Figma-like drag-and-drop workflow editor using ReactFlow |
| Multi-LLM Support | OpenAI, Claude, Gemini, Mistral, xAI, Ollama (local) |
| 1,000+ Integrations | Slack, Gmail, Supabase, Pinecone, and more |
| Deployment Options | API endpoint, scheduled task, chatbot, webhook trigger |
| Knowledge Bases | Upload docs to vector store for RAG-grounded agents |
| Copilot | AI assistant that generates nodes and fixes errors from natural language |
| Tables & Docs | Built-in structured data and document management |
| Self-Hosting | Docker Compose, devcontainer, or npx simstudio one-liner |
| CLI/SDK/MCP | Programmatic access across all tiers |
Deployment Modes
Workflows can be triggered in multiple ways:
- Manual — Run from the UI
- API — Deploy as a REST endpoint
- Scheduled — Cron-style periodic execution
- Webhook — React to incoming events (e.g., Slack messages)
- Chat — Deploy as a standalone chat interface
Tech Stack
| Layer | Technology |
|---|---|
| Framework | Next.js (App Router) |
| Runtime | Bun |
| Database | PostgreSQL + pgvector (Drizzle ORM) |
| Auth | Better Auth |
| UI | Shadcn + Tailwind CSS |
| State | Zustand |
| Flow Editor | ReactFlow |
| Realtime | Socket.io |
| Background Jobs | Trigger.dev |
| Code Execution | E2B (sandboxed) |
| Monorepo | Turborepo |
Pricing
| Tier | Price | Credits | Storage | Key Limits |
|---|---|---|---|---|
| Free | $0 | 1,000 (trial) | 5GB | 3 tables, 5 min execution, 7-day logs |
| Pro | $25/mo | 6,000/mo + 50/day | 50GB | 25 tables, 50 min execution, 150 runs/min |
| Max | $100/mo | 25,000/mo + 200/day | 500GB | 25 tables, 50 min execution, 300 runs/min |
| Enterprise | Custom | Custom | Custom | SSO/SCIM, SOC2/HIPAA, self-hosting, dedicated support |
Funding & Investors
$7M Series A — November 2025
| Role | Investor |
|---|---|
| Lead | Standard Capital |
| Institutional | Perplexity Fund, SV Angel, Y Combinator, Liquid 2 Ventures, Twenty Two Ventures |
| Notable Angels | Paul Graham, Paul Bucheit, Ali Rowghani, Kulveer Taggar, Kaz Nejatian, Chris Howard |
The investor list is remarkable for a $7M round — Paul Graham personally investing alongside the Perplexity Fund signals strong conviction from people who see a lot of AI companies.
Traction & Community
| Metric | Value |
|---|---|
| GitHub Stars | 27,125 |
| GitHub Forks | 3,443 |
| Open Issues | 219 |
| Claimed Users | 100,000+ builders |
| Repo Age | ~14 months (Jan 2025) |
| Last Push | Active (daily commits) |
27K stars in 14 months is strong growth. For context, n8n has ~63K stars but has been around since 2019. Sim's trajectory suggests it could challenge established workflow tools within 1-2 years if growth continues.
The 100K+ builder claim should be taken with standard startup skepticism — this likely includes free tier signups and self-hosted instances, not paying customers.
Competitive Landscape
Sim competes across two axes: visual workflow builders and agent frameworks.
| Tool | Type | AI-Native? | Open Source? | Stars | Differentiator |
|---|---|---|---|---|---|
| Sim.ai | Visual builder | Yes | Yes (Apache 2.0) | 27K | Figma-like canvas, LLM-first design |
| n8n | Workflow automation | Partial | Yes (Fair-code) | ~63K | Broadest integrations, general automation |
| Langflow | Visual builder | Yes | Yes (MIT) | ~50K | LangChain-native, Python ecosystem |
| Flowise | Visual builder | Yes | Yes (Apache 2.0) | ~35K | LangChain + chatbot focus |
| Dify | AI app builder | Yes | Yes (Apache 2.0) | ~70K | RAG + prompt IDE + app hosting |
| CrewAI | Code framework | Yes | Yes (MIT) | ~30K | Multi-agent, code-first |
| AutoGen | Code framework | Yes | Yes (CC-BY-4.0) | ~40K | Microsoft-backed, research-oriented |
Where Sim Wins
- UX polish — The Figma analogy is real. The visual canvas is cleaner than Langflow/Flowise
- Modern stack — Bun + Next.js + Turborepo is a developer-friendly foundation
- Copilot integration — Natural language flow generation is a strong differentiator
- Self-hosting simplicity —
npx simstudioone-liner is hard to beat
Where Sim Is Weak
- Maturity — 14 months old, ~5 employees. Enterprise features (SOC2, HIPAA) are listed but likely early
- Ecosystem depth — n8n's 400+ integration nodes and battle-tested production deployments are years ahead
- Star-to-revenue gap — 27K stars with a tiny team and credit-based pricing suggests monetization is nascent
- Platform assumptions — As Reddit users note, Sim "trades flexibility for platform assumptions"
Analysis
Strengths
Premium domain + strong brand. sim.ai is a killer domain for an agent workflow company. Combined with clean design and a clear value prop, the brand punches above its weight class.
YC + PG personal investment. Paul Graham investing personally in a $7M round is a strong signal. The Perplexity Fund participating suggests other AI companies see Sim as infrastructure, not competition.
Open-source growth flywheel. 27K stars creates a self-reinforcing cycle: community contributions, hiring pipeline, trust signal for enterprises. Apache 2.0 licensing removes adoption friction.
Right time, right category. Agentic AI workflows are the hottest category in enterprise software right now. Sim is positioned at the intersection of the visual builder trend and the agent framework explosion.
Risks
Tiny team in a crowded market. 5 employees competing against n8n (~200 people), Dify (well-funded), and the entire LangChain ecosystem is a real challenge. The "moat" question is sharp — visual workflow UX is easy to replicate.
Credit-based monetization. The pricing model (1,000 credits free, 6,000 for $25/mo) is unproven. It's unclear what a "credit" maps to in actual agent runs, and whether the unit economics work.
Copilot dependency on managed service. Self-hosted instances need a Copilot API key from sim.ai — this creates a managed service dependency that could frustrate self-hosters.
Star inflation risk. The open-source AI space has a pattern of rapid star accumulation followed by low conversion. Stars ≠ production deployments.
Tembo Relevance
Sim is in the visual agent orchestration layer — adjacent to but not directly competing with Tembo's coding agent orchestration focus. However, Sim's growth signals how quickly the "agent builder" category is expanding. Worth monitoring as a potential integration partner or as a proxy for where the market is heading.
Bottom Line
Sim.ai is a well-positioned, well-funded open-source play in one of the fastest-growing categories in software. The YC pedigree, premium domain, Paul Graham backing, and 27K-star growth curve are all strong signals. The main question is whether a 5-person team can build enough depth to compete with much larger workflow platforms while the window is open. The next 12 months will tell — watch for enterprise customer logos, revenue metrics, and whether the community converts from tire-kickers to production users.
Verdict: Early but promising. The investor roster alone makes it worth watching.