Opportunities never happen. You need to create them.
I'm building CreateOS. The execution layer that lets anyone create, deploy, monetize, and scale apps and agents without DevOps expertise. My mission is to deliver the most simplified experience for customers across AI, blockchain, supply chain, legal, and SMBs. A unified AI execution layer.
I write here when I have something worth keeping, learning, and sharing.
on the Vercel AI supply-chain breach · read · /writing
/about
san francisco · memo 01
Co-Founder & CEO of NodeOps. Building CreateOS.
how I think
Revenue is the proof. Everything else is a narrative.
Ship before you're ready. Velocity forces clarity.
The hard problem is rarely the code. It's everything that comes after.
Build for the people who can't already build.
Tools either remove friction or add it. There is no neutral.
Distributed by default. Centralisation is the exception.
Bias toward action. Deliberation is a tax on momentum.
how I work
In public. With team. Always.
I'd rather hand someone the rough version they can criticise than describe the perfect one they can't see. The conversation that follows is the actual work.
Most of what looks like strategy is just constraints I haven't named yet. When I name them, the path opens.
I write to think. The post, the deck, the doc are all forms of me arguing with my own assumptions. If you want to know what I believe, read what I shipped.
The teams I trust most can name what they got wrong this quarter without flinching. That's the cultural test.
influences
people whose voice keeps me honest
Naval Ravikant. First principles. Terse truth-telling.
Sahil Bloom. Structured wisdom. Generosity in public.
ideas I keep returning to
AI belongs at the OS layer, not bolted on as a feature.
Revenue is the strongest form of proof.
Most strategy failures are constraint-naming failures.
Compound interest applies to learning and reputation, not just capital.
Build for the people who can't already build.
/speaking
san francisco · memo 02
talks · sessions · booking
what I speak on
enterprise AI adoption
AI-native deployment as the next infrastructure layer
founders must create opportunities every single day
shipping your products reliably at scale
revenue-first approach for anything you do
agent orchestration and execution
a truly unified operating system
build it now
A live-build series. We sit down with someone who has held onto an idea because they didn't think they were the right person to build it. We ask one question. Then we build it with them, on CreateOS, in a single session. From Bangalore to Rome to Delhi.
ep 1
Kalash Vasaniya · Justref · affiliate platform · Bangalore · watch · case study
ep 2
Stefania · TACBoard · custom productivity tool · Rome · watch · case study
ep 3
Pratyush · Agent Blaster · AI agent benchmark · Delhi · watch · case study
past talks & interviews
06 may 26
CreateOS Lounge · live with Lucas (BNB Chain) on reducing builder friction · watch
30 jun 25
Blockchain Interviews w/ Ashton Addison · on DePIN, DeFi, and AI · watch
26 jun 25
DePINed Podcast w/ Tom Trowbridge · the DePIN orchestration thesis · watch
27 mar 25
SOL DePIN Summit Bangalore · keynote on orchestration as the foundation of DePIN · watch
14 mar 25
DePIN Day Denver · hardware-free DePIN compute · watch
07 mar 25
DePIN Day Hong Kong · keynote on the orchestration layer thesis · watch
15 oct 24
Korea Blockchain Week · simplifying node operations across major chains · watch
Naman Kabra is the Co-Founder and CEO of NodeOps, building CreateOS — the enterprise execution layer for AI-powered software. NodeOps operates one of the largest distributed compute networks in production, spanning tens of thousands of nodes across dozens of chains. CreateOS launched #1 Product of the Day on Product Hunt in February 2026.
150 words
Naman Kabra is the Co-Founder and CEO of NodeOps, where he leads CreateOS — the enterprise execution layer for AI-powered software. CreateOS lets enterprises and builders build, deploy, operate, and monetise AI applications and agents on a single platform, with zero cloud lock-in and no DevOps overhead.
NodeOps operates one of the largest distributed compute networks in production, spanning tens of thousands of nodes across 44+ chains, with enterprise customers across AI infrastructure, DePIN, and validator operations. CreateOS launched on top of that infrastructure and hit #1 Product of the Day on Product Hunt in February 2026.
Before NodeOps, Naman led enterprise partnerships at Metasky (Sequoia-backed), co-founded and exited a social payments startup, contributed to validator infrastructure at Persistence in the Cosmos ecosystem, and led a Center of Excellence at Robert Bosch. He has been researching blockchain since 2017.
300 words
Naman Kabra is the Co-Founder and CEO of NodeOps, the distributed cloud infrastructure company powering the next generation of AI-native software. At NodeOps, Naman leads CreateOS — the enterprise execution layer where AI-powered applications and agents are built, deployed, operated, and monetised on a single platform, with no cloud lock-in and no DevOps overhead.
CreateOS launched in February 2026 and hit #1 Product of the Day on Product Hunt on launch day. It runs on the NodeOps infrastructure layer: tens of thousands of nodes under management across 44+ chains, with enterprise customers across AI infrastructure, DePIN, and validator operations. CreateOS is integrated natively with the major AI development environments — Claude Code, Cursor, ChatGPT, Gemini CLI, Windsurf, and OpenClaw — so enterprises and builders move from prompt to production in minutes.
NodeOps is backed by L1D, Blockchain Founders Fund, and Finality Capital Partners, with angel participation from Sandeep Nailwal (Polygon), JD Kanani (Mozaik), and Richard Ma (Quantstamp), alongside strategic investor-clients including Arbitrum and 0G Labs.
Before NodeOps, Naman led enterprise partnerships at Metasky, a Sequoia- and Woodstock-backed consumer Web3 company, where he activated Web3 experiences for global brands. He co-founded and exited a social payments startup, contributed to validator infrastructure work at Persistence in the Cosmos ecosystem, and previously led a Center of Excellence at Robert Bosch's global engineering organisation. Naman has been researching blockchain since 2017, with published work on blockchain frameworks for financial institutions and IoT networks. He holds a BTech from Nirma University, Ahmedabad.
book me
naman-at-nodeops-dot-xyz
/work
san francisco · memo 03
career, role by role
I started in blockchain research in 2017. Since then, every role has been about turning protocol-level ideas into something a real user could touch.
now
NodeOps · Co-Founder & CEO · 2023–present
Distributed cloud infrastructure. The orchestration layer underneath AI-native deployment. Tens of thousands of nodes across dozens of chains. CreateOS shipped on top in 2026.
before NodeOps
Metasky · Business Development
Consumer Web3 startup backed by Sequoia and Woodstock. Led enterprise partnerships activating Web3 experiences for global brands.
Social payments startup · Co-Founder
Shipped a payments product. Grew it. Sold it.
Persistence
Cosmos ecosystem. Protocol-level work on validator infrastructure.
Robert Bosch · Center of Excellence Lead
Innovation work at the global engineering organization. Where I learned what enterprise actually expects from software.
education
BTech · Nirma University, Ahmedabad · Institute of Technology.
advising
Early-stage builders working on AI infrastructure, asset tokenization, GameFi, and consumer engagement.
/nodeops
san francisco · memo 04
the orchestration layer · 2023 to now
why we started
In 2023, every Web3 team was reinventing the same infrastructure. Running a node was a research project. Operating across chains was a full-time job. The teams building real products were spending more time on DevOps than on the product itself.
The bet: orchestration should be the boring middle layer, not the bottleneck.
the thesis
Decentralized compute and AI workloads are converging. Both need a coordination layer that is permissionless, verifiable, and abstracted away from the people building on top. NodeOps is that layer.
The architecture splits cleanly. NodeOps is the orchestration layer. It decides which node runs what, where, with what guarantees. CreateOS, which we shipped on top in 2026, is the execution layer. Orchestration below. Execution above.
co-founders
Pratik Balar built the infrastructure. From the first node deployment to the current network, he is the reason it actually works.
Jagdish Chudasama designed CreateOS and runs developer education. His YouTube channel teaches builders how to ship on the platform.
what we have built
A decentralized compute network spanning tens of thousands of nodes across dozens of chains. Validator operations. Multi-chain RPC. Containerized infrastructure for AI workloads. Generating revenue since 2023 from real customers. Seed funding closed.
partnerships
Working with teams across the AI and DePIN ecosystems. Sarvam for Indic language workloads. Fluence for permissionless cloud. Scaler Innovation Lab as an infrastructure execution partner. The list grows.
where we are now
CreateOS launched on Product Hunt in February 2026 and hit #1 on launch day. The early case studies are real builders shipping real products: Justref, TACBoard, Agent Blaster, Riyasat, AnyWhereGYM, Distraction Tracker, and the industrial AI pilot for Indian textiles.
NodeOps continues underneath. Same thesis, more nodes, more chains, more workloads. Orchestration is no longer the bottleneck.
/createos
san francisco · memo 05
the execution layer for AI
the deployment wall
Most AI-built products never reach production. Not because of the code. AI writes code faster than ever.
The wall is everything that comes after: infrastructure configuration, CI/CD pipelines, SSL provisioning, environment setup. Momentum dies exactly when the product is ready to ship.
That gap is what CreateOS exists to remove.
what it is
CreateOS is the execution layer for AI. Where AI apps and agents actually run in production. Build, run, scale, and monetize. One place.
A builder describes what they want. CreateOS deploys it in minutes. No DevOps knowledge. No tool fragmentation. No infrastructure overhead. From day one, the app is live and can be monetized.
why we could build this
CreateOS doesn't exist without NodeOps. Before CreateOS, NodeOps built and operated distributed cloud infrastructure across tens of thousands of nodes and dozens of chains. That infrastructure is what CreateOS runs on.
It's also why CreateOS can offer blockchain-native AI skills, zero cloud dependency, and a structural cost position no developer-tools company can match. The foundation predates the product by two years.
The architecture splits cleanly. NodeOps is the orchestration layer. It decides which node runs what. CreateOS is the execution layer. It's where work actually runs after the orchestrator routes it. Orchestration below, execution above.
who it's for
Builders who can describe what they want to ship but don't want to become DevOps engineers in the process. Non-technical founders. Indie hackers. Teams shipping AI apps that need to handle real traffic, real users, real revenue. Not vibe-coding experiments. Production infrastructure.
user stories
Stefania is a designer in Rome with no coding background. She built TACBoard on CreateOS, a custom productivity tool that replaced six separate apps. She didn't learn to code. She learned to describe what she wanted. Watch the session.
A textile facility in Maharashtra ran their CCTV compliance system on CreateOS. The 75-day pilot extended across four Indian states. Read the case study.
These are not edge cases. They are the median user.
/ideas
san francisco
ideas I'm currently working on
Working notes. Not all of them will ship. The ones I cannot stop thinking about.
01 · A true operating system for agentic products and enterprises
Not another agent framework wrapped around a chat box. The base layer where agentic products run, where enterprise workflows orchestrate them, where the OS itself understands what the agents are doing and why.
02 · A voice-to-voice agent for quick commerce and insta-help
Built for Tier-2 cities in India. Voice in, voice out. No app. No typing. The phone call as the interface. The next billion users will not type their way to commerce.
03 · ipo.new
A signal protocol for ideas. Drop your napkin idea. Let agents and humans validate it before you write a single line of code.
04 · A desk-placed mini-robot to work with you
Your agents with a soul, a brain, a voice, an identity. Quite literally on your desk. The presence layer for the agents you already run.
05 · Tool-call monetization and affiliation layer
Agents make API calls all day. Someone should be earning a cut. A layer for actually getting your digital work paid. Affiliation as infrastructure, not afterthought.
/writing
san francisco · archive
essays · long-form · founder thinking
recent
The Vercel breach is not a Vercel problem
April 28, 2026 · On the first major AI supply-chain breach since AI tools went mainstream. The vector matters more than the target. What every team using OAuth-connected AI tools should do in the next 72 hours, and what the architectural lesson is for the next two years. Read on Substack.
archive
Cointelegraph columns
Pieces on Web3 infrastructure, AppChains, NFTs, and onboarding users from Web2 to Web3. Full archive.
elsewhere
Op-eds, profiles, and contributor pieces in trade publications. Aggregated under /press.
what I write about
AI-native deployment as the next infrastructure layer
distributed cloud, orchestration, and the case against single-vendor consolidation
building global infrastructure from outside Silicon Valley
founder operating principles tested against real revenue
/research
san francisco · memo 06
papers · current interests · academic background
current interests
blockchain
distributed ledger technology
memory layer for agents
agent harness
published
MudraChain: Blockchain-Based Framework for Automated Cheque Clearance in Financial Institutions
August 2019 · with Pronaya Bhattacharya, Sudeep Tanwar, Sudhanshu Tyagi · paper
A Systematic Framework for State Channel Protocols Identification for Blockchain-Based IoT Networks and Applications
June 2020 · with Harsh Mashru, Krishnan Mohan · paper
background
BTech · Nirma University, Ahmedabad · Institute of Technology. Started researching blockchain in 2017, contributing to the open-source research community before founding NodeOps.
/press
san francisco · archive
coverage · interviews · profiles
profiles & interviews
CEOWORLD Magazine · March 2025
CEO Spotlight on onboarding users from Web2 to Web3 and the future of decentralized infrastructure. Read.
Korea IT Times · June 2025
Exclusive interview on Web3 infrastructure innovation and NodeOps' global expansion. Read.
Industrial AI pilot for Indian textiles · February 2026
GlobeNewswire release picked up by Cryptopolitan, TechBullion, Manila Times, StreetInsider, and FinancialContent. Press release · CreateOS case study.
