Skip to content
Paris, France · 48.8566°N

Thăng Long — 昇龍 — the ascending dragon

LongTang

  Product Engineer

I work at the collision of design and engineering — where most people pick a side.

View workGet in touch

Scroll — ink to grid

01In production · inside FPT Software France

ProspectIQ

An AI-powered B2B prospecting platform, used daily inside FPT — rebuilt to survive the death of its core data vendor.

  1. 01The scraping data source died (Proxycurl, post-LinkedIn lawsuit). A live, depended-on tool was about to break.
  2. 02Multi-provider with fallback — no single vendor can take the system down. A queue + retry layer — no API ceiling can break the pipeline.
  3. 03Then Proxycurl actually shut down for good — and the tool kept running. The thesis paid off in reality.

95%

cheaper than per-seat prospecting tools

Build for the failure case first — vendors will die on you.
02Live · no sign-up · target7.io

Target7

An essay grader whose scores land as close to a human examiner's as two examiners land to each other.

  1. 01Deterministic at temperature 0 — the same essay gets the same score, every time.
  2. 02Calibrated with examiner re-anchoring + isotonic regression; DeepSeek + Gemini two-pass ensemble.
  3. 03On held-out examiner-marked essays, agreement reached |Δ| ≈ 0.45 — near the inter-rater agreement between two humans.

99%

cheaper per grading than a human tutor

Knowing a metric's limits is part of the work.
03Range, proven

Four more shipped — different domains, same instinct.

01Open the tools

PDFForge

26 PDF tools where files never touch a server — privacy is the architecture.

02Private demo — on request

Skyhop

Paris ↔ Vietnam flight finder tagging every route by airspace safety. The failure-first thesis, again.

03Walkthrough on request

CANAL+

UX audit delivered as a full design system: tokens, components, docs.

04Case study on request

Orange Brain

A learning PWA built as a complete design system, Nothing-OS aesthetic, end to end.

04The collision

Most people pick a side — design or engineering. I never could. I build at the seam between them, because that's where the interesting problems live. I can take something from a user interview to a deployed, AI-powered product without a hand-off losing the thread.

Before the AI work I ran cross-regional UX research at FPT — 15 participants across Europe and Asia. And my first product was growth: four years running digital for my family's restaurant in Germany — 3.8 to 4.5 stars, 1,400+ reviews. Shipping means moving a real metric.

  • 01Trilingual — VN · FR · EN
  • 02Master's UX Design — Efrei Paris / Panthéon-Assas
  • 03BSc Computer Science, Honours — Université Paris Cité
  • 04Alternance — FPT Software France