This is the traditional view of a career better experienced as a system: THE CONVERGENCE →
Use your browser’s Print → Save as PDF for a downloadable copy.
Majid Asghari Tabrizi
Product Lead — AI — Systems · AI-native organization design
quantiviq.xyz · majid@quantiviq.xyz · github.com/MajidAsghariTabrizi · linkedin.com/in/majid-asghari
Thesis
One career, deeper layers: how software works → how products create value → how teams create products → how organizations operate → how complex systems make decisions → how AI can act → how AI can learn → how human + AI organizations should operate. Not a sequence of titles — a sequence of increasingly harder questions about reality, decisions, autonomy and learning.
Career layers
| 2008–2014 | TECHNOLOGY | How software works — and how it survives contact with production. |
| 2014–2018 | PRODUCT | How products create value — for whom, and what changes if you build it. |
| 2018–2020 | ORGANIZATION | How teams create products — the system around the code. |
| 2020–2023 | MARKETPLACE SYSTEMS | How organizations operate — every decision propagates. |
| 2023–2026 | DECISION SYSTEMS | How complex systems make decisions — under uncertainty, with consequences. |
| 2025 | AI INFRASTRUCTURE | Intelligence became infrastructure — routed, ranked, budgeted. |
| 2025–2026 | AGENT RUNTIME | From model to worker — governed execution, recovery, verification. |
| 2026 | LEARNING SYSTEMS | What if the system didn’t start from zero every time? |
| NOW | LEARNING SYSTEMS | AI-native operating systems — the convergence of every layer. |
Employers (owner-verified list): SnappMarket · Okala · Alibaba Travels Co. · SportMob · Daapapp · Infinite8. Role details available on request.
Certification: Professional Scrum Master I (PSM I).
Selected systems (public proof)
Phoenix — Retired 2026-09-17 — disciplined shutdown, measured negative result
Production-grade, fail-closed financial intelligence and execution system for Arbitrum — retired after measurement, with the negative result kept public.
github.com/MajidAsghariTabrizi/anti-gravity-phoenix-v4
Smart Trader — Archived · data frozen 2026-09-23
A decision-intelligence trading system: market structure, behavior signals, and explicit risk gates.
github.com/MajidAsghariTabrizi/smart-trader
Free Best Router — MIT · v0.1.0 released
One OpenAI-compatible endpoint that automatically discovers, ranks, and routes to the best available free AI model.
github.com/MajidAsghariTabrizi/free-best-router
Universal Engineering Agent — MIT · reference implementation
A profile-agnostic, runnable, MIT-licensed reference implementation of an engineering-agent operating-kernel contract.
github.com/MajidAsghariTabrizi/universal-engineering-agent
How I work
- Reality over architecture; evidence over confidence; no edge, no bet.
- Abstention is a valid decision; fail closed when downside is asymmetric.
- Cheapest experiment before machine; separate intelligence from authority.
- Learning must preserve provenance; technical difficulty is not commercial validation.