What is AI-native development?
AI-native development is an operating model in which a team redesigns its software process around AI instead of adding AI to the old one; Tessl’s phrase is “built-in, not bolted on”. Guy Podjarny introduced “AI Native Software Development” at Tessl’s launch on 6 July 2024. It describes how a team works, not one developer’s habit, and says nothing about autonomy.
Origin Guy Podjarny introduced the name “AI Native Software Development” in Tessl’s launch post (opens in a new tab) on 6 July 2024, and Patrick Debois later described four patterns of it in “The 4 patterns of AI Native Dev” (opens in a new tab) on 19 March 2025, without having coined it.
- In Polish
- Programowanie AI-native
- Closest ladder level
- All levels, free guide
- Also searched as
- AI native development, AI Native Dev, AI-native software development, AI-native engineering
Then $19.99 a month or $99.99 a year. Cancel from your account page. 30-day refund on a first purchase.
Why it matters.
The label is easy to claim and hard to test. Tessl set it against AI-assisted development, “the same dev workflow, but enhanced with AI”, and Nearform’s Cian Clarke warns: “Many companies think enabling Copilot means they’re doing AI-native engineering.” The term also says nothing about autonomy, since Dan Shapiro puts “90% of ‘AI-native’ developers” at Level 2. Ask what changed in the process instead: who writes the first pass, which artifact each stage commits, and who owns each gate.
AI-native development vs vibe coding
AI-native describes a team’s process and vibe coding describes one person’s review habit, so an AI-native team can vibe code a prototype and still ship through gates.
| Question | Vibe coding | AI-native development |
|---|---|---|
| What it names | One person’s way of working: accept the model’s output without reading it | A team’s operating model: the process is built around AI |
| Who writes the code | The model | Agents write the first pass; the engineer is “the reviewer, editor, and source of direction” (OpenAI’s guide) |
| Who checks it | Nobody reads the diff; the human judges behaviour by eye | An agent can do the first review; engineers “own the final review and merge process” (OpenAI’s guide) |
| What stops the loop | The human decides it works | Not defined by the term |
| Where the name came from | Andrej Karpathy, post on X, 2 February 2025 | Guy Podjarny, Tessl’s launch post, 6 July 2024; Patrick Debois described four patterns on 19 March 2025 |
| Where it breaks | Anything others maintain; Karpathy called it fine for “throwaway weekend projects” | The label proves little: “Many companies think enabling Copilot means they’re doing AI-native engineering.” (Cian Clarke, 6 January 2026) |
Patrick Debois’s four AI-native patterns
Debois offered a “first pass at four AI Native Dev patterns that are currently emerging”, each a shift in a developer’s role.
| Pattern | What changes |
|---|---|
| 1 · From producer to manager | AI produces more code, so you spend more time reviewing and deciding; with autonomous agents, “we’re essentially becoming managers of agent development teams” |
| 2 · From implementation to intent | You supply intent and richer requirements instead of the implementation, and when output is wrong, “we simply update the specifications” |
| 3 · From delivery to discovery | Experiments get cheap because AI can prototype ideas and options, so teams compare alternatives and run experiments through their CI/CD pipelines |
| 4 · From content to knowledge | Documentation goes stale, and AI gives a reason to share knowledge, which benefits colleagues and the AI systems themselves |
Questions about AI-native development.
How is AI-native development different from vibe coding?
Vibe coding is one person’s review habit: accepting AI-written code without reading it. AI-native development is a property of a team’s process. Our reading: an AI-native team can vibe code a prototype and still ship through gates, which fits Debois’s “from delivery to discovery” pattern. Cian Clarke of Nearform adds that spec-driven development “removes the ceiling that limits pure vibe coding”.
Who coined AI-native development?
Nobody is established as the coiner. Tessl’s Guy Podjarny introduced “AI Native Software Development” as a paradigm on 6 July 2024. Patrick Debois, whom Tessl credits with coining DevOps, described four patterns of it on 19 March 2025, but did not coin the term. Gartner adopted “AI-native software engineering” in a press release of 1 July 2025.
What is the difference between AI-native and AI-assisted development?
Tessl drew the line at its launch: AI-assisted development is “the same dev workflow, but enhanced with AI”, while AI-native development is “built-in, not bolted on”. AWS draws a similar line, describing AI-assisted development as “where AI enhances specific tasks like documentation, code completion, and testing”, and offers AI-DLC as a third way.
Does being AI-native mean a team is high on the autonomy ladder?
No, the term says nothing about autonomy. Dan Shapiro writes that Level 2 “is where 90% of ‘AI-native’ developers are living right now”. This site’s ladder gives a level to a loop, not to a team: a dependency-bump loop can run at Level 4 while feature work beside it sits at Level 2.
The vocabulary of AI-driven development
Every name below is defined against vibe coding: who writes the code, who checks it, and what stops the loop.
Read the long-form guide: agentic engineering vs vibe coding
Sources.
The primary sources outside this site that this page relies on.
- Announcing Tessl, the AI Native Development Startup (opens in a new tab) Guy Podjarny, Tessl
- The 4 patterns of AI Native Dev - Overview (opens in a new tab) Patrick Debois, Tessl
- Gartner Identifies the Top Strategic Trends in Software Engineering for 2025 and Beyond (opens in a new tab) Gartner
- Building an AI-Native Engineering Team (opens in a new tab) OpenAI updated
- From Vibe Coding to Spec-Driven Development (opens in a new tab) Cian Clarke, Tessl
- The Five Levels: from Spicy Autocomplete to the Dark Factory (opens in a new tab) Dan Shapiro
- AI-Driven Development Life Cycle: Reimagining Software Engineering (opens in a new tab) Raja SP, AWS DevOps & Developer Productivity Blog
Keep reading.
The guides that go deeper, and the terms and comparisons next to this one.
In the docs
- The full A-Z glossarySubscription
- The autonomy ladder: which level is your workflow at?Free
- Vibe coding vs agentic engineering: 21 terms comparedFree
- The AI-native software development lifecycleSubscription
- Strategy: AI-native engineering for business leadersSubscription
- AI-native delivery for software houses and agenciesSubscription
- One map: the autonomy ladder, the lifecycle and the factory stationsFree
- Level 1–2: Assisted and Paired CodingFree
Related terms
Comparisons
Read the guides in the same words.
Open every guide with the 7-day free trial. Each term is defined once here, and the guides use it the same way.
Then $19.99 a month or $99.99 a year. Cancel from your account page. 30-day refund on a first purchase.