

Mael
@trymael
- CTO & Co-founder Wenoble
- Webflow Certified Partner
AI-native software engineer. I build software at the speed of AI, with the standards of real engineering.
Selected work
View allEngineer first. AI-native by default.
From idea to production, end to end.
I'm Maël, a software engineer and co-founder of Wenoble, with four years of full-stack work behind me.
I use LLMs, agents and Claude Code to move faster, and I own every line that ships: architecture, review, tests, security.
I work across the stack, from TypeScript and Python to Postgres, Docker and Linux servers, and I care about what happens after launch.
- Years building software
- 4+
- Projects delivered
- 50+
- Products built
- 3
- Things still to build
- ∞
Contributions
The stack I build and ship with.
- ReactReact
Component-driven interfaces, from internal dashboards to product UIs, built to stay maintainable.
- Next.jsNext.js
My default for products: server rendering, routing and APIs in one codebase, ready for production.
- TypeScriptTypeScript
Strict types end to end. It's also what keeps AI-generated code honest and reviewable.
- Tailwind CSSTailwind CSS
Styling in the code, with a consistent design system instead of one-off CSS.
- WebflowWebflow
Marketing sites teams can edit themselves, with a clean structure and custom code where it counts.
- Node.jsNode.js
APIs, workers, scripts and integrations on the server: the glue of most products.
- ExpressExpress
Lightweight APIs, webhooks and services when a full framework would be overkill.
- NestJSNestJS
Structured backends for larger APIs: modules, dependency injection, testable architecture.
- PostgreSQLPostgreSQL
The database I trust: relational, transactional, and still flexible with JSONB.
- MongoDBMongoDB
Document storage when data is deeply nested or its shape changes often.
- SupabaseSupabase
Postgres with auth, storage and access policies built in. Fast to start, solid to scale.
- ClaudeClaude
My main model: through the API for agents and RAG, through Claude Code to build faster.
- OpenAIOpenAI
GPT models and embeddings for generation, classification and semantic search.
- MCPMCP
Model Context Protocol: giving agents controlled access to real tools, APIs and data.
- VercelVercel
A preview on every branch and zero-config production deploys for Next.js apps.
- CloudflareCloudflare
DNS, CDN, security and edge workers in front of everything I put online.
- DockerDocker
Reproducible environments, from local development to VPS and production containers.
- GitHubGitHub
Code review, CI/CD with Actions and issues to keep every release under control.
Services
View allBlog
View all
How much does it cost to develop a custom SaaS in 2026?
In 2026, integrating AI features into a SaaS has become the standard in many industries. A simple chatbot via the OpenAI API costs relatively little. A RAG architecture or autonomous agents can significantly increase the development budget.
- Saas
- AI

Grokbots: Understanding xAI's Autonomous AI Agents Explained
Mid-2026, xAI made a bold move. With the launch of Grokbots, Elon Musk's company is no longer just offering a chatbot it's deploying autonomous AI agents capable of working on your behalf, in the background, without you having to monitor every step.
- AI





