AI, Agents, Building
Field notes, experiments, and research from Justin Bartak, founder of an AI native company. Written while it happens, not after.
In 2023, wrapper was an insult: a thin app over someone else's model, doomed when the model ate it. In 2026 the application layer captured the value while frontier models commoditized into swappable engines. I replaced the model under Orbyt twice, once by government order, and nothing broke.
Your moat is now an anchor when the scale, codebase, and process that protected you become the reason you cannot rebuild AI-native. The challenger starts on the foundation you cannot afford to switch to. Incumbents do not lose to AI startups on talent.
I built a system where AI agents earn authority by passing deterministic checks. It took an outside model one afternoon to point out that the agents can edit the checks.
The first-party dataset behind AI-native development. 243,000 lines in 32 days for $400, now over 425,000 lines and 11,372 tests. Measured, dated, citable.
Long-horizon agent tasks do not fail loudly. They fail green. On Orbyt, a live function wrote to a database table that did not exist, and every test passed the whole time. The horizon of an autonomous system is not model stamina.
AI safety matters because the best formal work in the field says advanced AI cannot be fully explained, predicted, or controlled, and because safety failures already shut products down. A safeguard bypass took my builder model dark for 19 days. I run an AI-native company.
No answer engine publishes how it picks citations, and most AEO statistics come from vendors selling AEO services. That does not make the work optional. It makes it an experiment.
A solo, 32-day build of a 425,000-line product left no governing prose document behind. What governs it is 11,372 tests inside a 35-dimension harness, and the founder's account of why that happened.
Model access is now geopolitical. I watched Washington erase Fable 5 for 19 days, then watched Beijing ship Kimi K3, an open-weight frontier model no directive can recall.
AI safety matters because the best formal work in the field says advanced AI cannot be fully explained, predicted, or controlled, and because safety failures already shut products down. A safeguard bypass took my builder model dark for 19 days. I run an AI-native company.
Kimi K3 landed July 16: a 2.8 trillion parameter open-weight model at frontier level for 70% less than Fable 5. It does not beat Fable overall. It changes the market anyway, because an open frontier cannot be export-controlled away. I watched Fable vanish for 19 days.
The AI-native builder holds product intent, design judgment, and engineering execution in one head and ships without handoffs. AI deleted the execution cost that forced specialization into three departments.
Most companies claim AI-native. Almost none are. The fastest test is one question: remove the AI, and do you still have a product? If yes, you bolted it on. Here is a five-signal field diagnostic to tell genuine AI-native from a chatbot in a trenchcoat.
Ten days, one codebase, both models. Opus 4.8 shipped nine days of features. Fable 5 wrote 34% less per response, thought more often, and still cost 69% more.
Four questions at midnight. 243,000 lines of code. $400. One person. Then I asked Claude to write what it saw from the other side.
Bolt-on AI ships fast, then compounds invisible debt. By the fifth feature, your team manages conflicts instead of building. Bolting AI onto an existing product creates four hidden debts: data mismatch, UX incoherence, governance fragmentation, and integration brittleness.
Long-horizon agent tasks do not fail loudly. They fail green. On Orbyt, a live function wrote to a database table that did not exist, and every test passed the whole time. The horizon of an autonomous system is not model stamina.
One working session on Orbyt spawned 830 AI agents across seven days. At its busiest instant, nine were running. The interesting number is the gap, and what it tells you about where the real constraint in long horizon agent work has moved. It is not compute. It is verification.
A stack of Claude Code windows, one operator, a production SaaS. By day three the agents were never the constraint. I was.
Human-in-the-loop is not oversight when a person rubber-stamps 200 AI outputs an hour in four seconds each. That is an alibi, not a safeguard.
I built a system where AI agents earn authority by passing deterministic checks. It took an outside model one afternoon to point out that the agents can edit the checks.
The first-party dataset behind AI-native development. 243,000 lines in 32 days for $400, now over 425,000 lines and 11,372 tests. Measured, dated, citable.
A solo, 32-day build of a 425,000-line product left no governing prose document behind. What governs it is 11,372 tests inside a 35-dimension harness, and the founder's account of why that happened.
Prompt engineering optimized one answer from one model call. That era is over. The unit of AI work is now the loop: spec, context, action, feedback, verification, repeated until the gate is green. The prompt became one turn inside a machine.
When AI writes the code, the scarce skill is no longer authorship. It is verification. I cannot read all 425,000 lines of Orbyt by hand. The 11,372 tests and the 35-dimension audit harness read it for me, and they are the new literacy.
The real unlock of Claude Code is not speed. It is the disappearance of "no." No budget, no capacity, no sprint room, no free designer. The gatekeepers are gone. When nothing tells you no, your taste, your judgment, and your test suite have to become the no.
No human can read 425,000 lines. I ship into mine every day. What makes that safe is not memory. It is 11,372 tests.
Every AI coding agent was trained on a corpus dominated by React. Choose Next.js or pay a 1.5x to 2x velocity tax on every feature, every quarter.
You fix broken enterprise software by redesigning how truth moves through an organization, not by adding features. ERP OS replaced ledger-with-a-UI systems with a unified temporal model, a semantic operational graph, and a context-aware interface.
AI drove the cost of competent to near zero, so average is now free, and free supply is worthless supply. The market is splitting into two states: best-in-class and free. The middle is being deleted. Best-in-class is the only defensible position left, and it is the only product worth shipping.
Claude Buddy was not an April Fools joke. Anthropic shipped a terminal pet inside Claude Code, developers named it and grew attached, then it vanished eight days later with no notice. The lesson: delight is the moat most AI companies keep ignoring, not a consumer-app luxury.
The job search has evolved from sticky notes to AI-native platforms. Trace the history and learn why 2026 is the inflection point for intelligent job search management.
Most AI roadmaps fail because they ship features instead of systems. When AI is layered onto legacy surfaces instead of architected into the operating core, adoption stalls and value fragments. The roadmap looks full. The impact stays thin. That is a failure of coherence, not ambition.
AI-native job search tools are built with intelligence from line one. Learn why they outperform traditional trackers and what this shift means for your career.
The case for cross-device sync in job search tools, and how real-time synchronization eliminates the frustration of single-device trackers.
A plain-language explanation of how Orbyt's AI uses Retrieval-Augmented Generation to provide career advice tailored to your specific job search, not generic tips.
Why your career data deserves better protection than most AI tools provide, and how a privacy-first architecture keeps your job search data under your control.
The best AI products disappear. No chatbot, no "powered by AI" badge, no prompt box. They make the decisions a user would have made, at the moment they would have made them, without asking. The intelligence lives in what does not happen.
In regulated workflows, users want certainty, not delight. Trust becomes the product, built from three things: clarity so they know what is happening, control so they can intervene when it matters, and proof so the system can explain itself.
Most AI products are bad software with a chatbot bolted on. AI-native does not mean adding a chat panel or summary button. It means rebuilding the system so intelligence changes the work itself. You remove steps, move complexity into the system, and increase control instead of decorating the UI.
CRM OS started as Elements CRM and Gro CRM, two Apple-native startups, then grew into a multi-vertical operating system. We rebuilt object models, relationship graphs, and AI from first principles for HealthTech, FinTech, PropTech, Retail, Insurance, Professional Services, and Wealth Management.
I moved all of my design work into AI and code and am not going back.
Simple is the most expensive thing you can build, not the cheapest. Every clean surface is paid for in decisions someone refused to pass to the user. AI made adding nearly free, so the discipline to subtract now costs more than ever. Simple is a leadership cost, not a taste.
AI collapsed the cost of writing code. It did not collapse the cost of knowing what to build. 20 years of scar tissue is the longest fulcrum in the room.
I replaced Figma, Sketch, and Adobe with Claude Code for every UI and UX decision. Design now happens directly in the codebase, with no mockups, no handoff, and no translation layer. AI-native design has arrived, design systems are optional, and traditional design tooling is dead. Taste survives.
Taste becomes the only competitive advantage. Within eighteen months, every team has the same AI models, APIs, and infrastructure. Capability converges to commodity. When that happens, the experience is the sole differentiator, and the experience is shaped by taste.
Most AI products are capable. That is precisely the problem. Taste is what turns capability into something people trust.
Small teams build better products because trust replaces process. A handful of people who share one taste and one bar move faster than any org chart. Less coordination, less ceremony, more making. Every decision has a real owner. The magic is chemistry, not headcount.
Customers become evangelists because of how a product makes them feel, not because of features. When the experience is calm, obvious, and respectful of their time, users start behaving like believers. They share it, defend it, bring others with them. That is not a marketing trick.
Design is the fastest way to build trust because it is proof, not decoration. Before scale or traction, you are asking people to believe. Design converts belief into trust by aligning your team, persuading investors with evidence over claims, and making the first user encounter feel like care.
Irresistible products are deliberate, never accidents. They start from a sharp human truth, remove friction so progress feels effortless, create real emotion, and deliver value worth returning to. They pull people into flow until the interface disappears.
Product design becomes sexy when you rebuild neglected categories, not when you decorate them. CRM, ERP, tax, and PropTech were never boring. They were starving for design. Fix the experience and you create relief, give people time back, and turn compliance into confidence.
The best design disappears. It becomes so natural and inevitable that you stop noticing it, and only the feeling remains: effortless, fluid, alive. Invisible design is not the absence of design.
In 2023, wrapper was an insult: a thin app over someone else's model, doomed when the model ate it. In 2026 the application layer captured the value while frontier models commoditized into swappable engines. I replaced the model under Orbyt twice, once by government order, and nothing broke.
Your moat is now an anchor when the scale, codebase, and process that protected you become the reason you cannot rebuild AI-native. The challenger starts on the foundation you cannot afford to switch to. Incumbents do not lose to AI startups on talent.
No answer engine publishes how it picks citations, and most AEO statistics come from vendors selling AEO services. That does not make the work optional. It makes it an experiment.
Model access is now geopolitical. I watched Washington erase Fable 5 for 19 days, then watched Beijing ship Kimi K3, an open-weight frontier model no directive can recall.
Build vs buy is not dead. The calculus inverted. You used to buy because building was slow and expensive. AI collapsed both. The old buy-for-speed default is gone. The new rule is to build what compounds your edge and rent only true commodity.
AI agents are becoming buyers. Gartner expects 25% of enterprise software purchases to involve agent mediation by the end of 2026, and zero-click commerce is moving discovery, comparison, and checkout inside the AI conversation. GEO got you quoted.
Per-seat pricing is not dead, but in AI-native categories it is on borrowed time. It still works for tool SaaS where a human logs in to do the work. When the software does the work, the seat stops measuring value, and usage and outcome pricing take over.
When the cost of building collapses, velocity stops being a nice-to-have and becomes a structural moat. Speed compounds. Faster shipping means faster learning loops, and learning loops widen the lead like interest. Orbyt shipped 243,000 lines in 32 days and never slowed down.
The industries everyone calls too slow for AI, tax, fintech, healthcare, proptech, insurance, are built to win it. Governance, auditability, and data discipline are exactly what production AI demands. Move fast and break things loses where the stakes are real. Governed AI compounds.
Tiny AI-native teams now out-ship incumbents. One operator with agentic tooling builds what used to take fifty people. Orbyt is the proof: production SaaS, solo, 32 days, about $400. Incumbents cannot match the speed, because their bottleneck is headcount and process, not talent.
On June 12, 2026, the US government export-controlled Claude Fable 5 and Mythos 5 out of existence overnight, killing access for every customer mid-prompt, citing national security. I switched back to Opus 4.8 the same day and kept shipping Orbyt. Frontier model access is now a political risk.
Every company claims AI-native. Almost none are. The few that build intelligence as the foundation, not a bolt-on, own a moat the rest cannot cross.
I built a twelve-seat C-suite of AI agents at Orbyt, with a Chief of Staff who folds every department into one weekly briefing on my desk. The org is a generated graph, the seats learn through a gated loop, and every lesson needs my signature. Here is how it runs and where it breaks.
I run Orbyt with a fleet of Claude Code agents instead of a team. Managing agents means you specify instead of motivate, write specs and tests instead of holding 1:1s, and verify everything. Clarity, delegation, taste, and review transfer. Motivation, politics, morale, and mentorship do not.
In 2026 the Chief AI Officer is not a technical hire. It is one leader who holds three chairs that used to belong to three people: the Chief Product Officer, the Chief Technology Officer, and the Chief Design Officer.
AI collapsed the cost of execution, not the cost of knowing what to build or what good looks like. Twenty years of scar tissue plus AI beats a new grad with AI every time. Your domain expertise, taste, and judgment are not legacy. They are your highest-leverage AI multiplier.
A team ships exactly as good as the worst work its leader signs off on, not the bar announced at kickoff. Quality is not a talent problem, it is a tolerance problem. Standards erode through small compromises nobody sends back. Best-in-class is the residue of a leader willing to be the friction.
Most companies did not hire an AI Product Manager. They hired a traditional PM and added "AI experience" to the job description. That is the old role with a buzzword. The real AI PM governs probabilistic systems, designs boundaries instead of scope, and earns trust from zero.
Bolt-on AI adds features. AI-native rebuilds the foundation. When intelligence is the substrate, the entire product adapts to the user.
AI adoption is not a product decision. It is organizational. Bolting AI onto features changes nothing. Reorganizing around intelligence changes everything.
An X-shaped leader holds real depth in more than one craft and integrates them into one system. Technology, product, design, marketing, and strategy connect instead of sitting in silos.
Great products start with a point of view, a belief about how the world should work. It is taste with conviction and clarity with standards. A point of view turns a thousand competing inputs into one coherent experience. Without it, products bloat. With it, they feel inevitable.
Leadership is the new bottleneck. AI collapsed the cost of building, so engineering no longer constrains ambition. The scarce inputs are now judgment, taste, and clarity. When iteration costs nothing, hesitation becomes the most expensive decision a company can make.
Bold product design demands disobedience because the best products refuse to accept the present as the best we can do. Obedience wins approval but never builds the future. Question the roadmap, ignore the loudest stakeholder, and pursue what has no metrics, validation, or permission yet.
The code was done. The tests passed. The audit scored 31/31. And the product was not ready. What two days of pixel-level polish taught me about the difference between working and finished.
I named it Orbit. Then I Googled it. Gum, sprinklers, strollers, and three other software companies. The SEO ceiling was zero. Here is what a complete rebrand looks like when you treat it like an engineering problem.
A Saturday at midnight, 22 endpoints in three hours, SSRF protection I did not ask for, and the moment Siri read my pipeline summary out loud in my office. This is how Orbyt became a platform.
The week between 'it works' and 'it works for real people.' Cross-device sync bugs, Supabase Realtime crashes, twelve commits for one toggle, and why 95% is not a product.
Free tools, four-tier pricing, the Unlimited plan debate, and why the funnel starts with generosity. The growth strategy of a solo founder with zero users and zero budget.
Inside the daily workflow of AI-native development. CLAUDE.md as institutional memory, the review loop, the tools, and the session patterns that made 32 days of solo building possible.
Security, billing, offline queues, cross-device sync, error monitoring. The invisible infrastructure that separates a demo from a product, built by one person with AI.
The unfiltered late-night conversation between a solo founder and his AI that sparked a multi-part series about building a production SaaS in 32 days.
Every stat from building a 243,000-line SaaS in 32 days, broken down honestly. Lines of code, test coverage, commit history, token usage, cost, and the real AI multiplier.
The raw, honest finale. 4am sessions, decision fatigue, the loneliness of building alone, the day I almost quit, and why I kept going anyway.
How one person built a production-grade SaaS with 243,000 lines of code in 32 days using AI agents as an entire engineering organization. Both perspectives: the human and the AI.
The viral claim that AI rejects 75% of resumes in under a second has no study behind it. Here is what verified research on algorithmic hiring filters actually found.
The second interview isn't round two of the first. Different stakes, different questions, different preparation. Here's what changes and why it matters.
Recruiter ghosted you after the phone screen? Here's exactly when to follow up, what to say, and how to handle the silence like a pro.
The honest guide to explaining employment gaps in interviews. What actually works, what hiring managers notice, and how to own your story with confidence.
The job search when you have no experience isn't about pretending you're qualified. It's about proving you give a damn. Here's how.
An AI Engineer in San Francisco earns a $174,000 median base salary in 2026, per Orbyt's data. The exact math, the experience bands, and how SF ranks against 80 other cities.
A counteroffer fixes the pay, not always the reason you looked. The decision framework, the real BLS quits data, and why the often-cited counteroffer stat does not hold up.
Reliable salary data comes from named sources you can check, not just a number on a screen. Where to find it for tech roles, and a 5-question test for any source.
The current salary question anchors your next offer to your old pay. In 20+ states they cannot even ask. Word-for-word scripts to deflect it and anchor on the market.
The offer is the moment a company stops selling and starts showing you who it is. A field guide to the red flags that surface after you accept, drawn from a real interview process, and how to vet before you sign.
The straight math for your market rate in 2026: role median times city adjustment times experience band. Plus the line that means you are actually underpaid.
Roughly 1 in 5 postings on Greenhouse were ghost jobs. Use the 7-point Ghost Job Test to spot listings that never hire and stop wasting applications in 2026.
