AI SaaS · Multi-Tenant · Multi-Agent · Finance & Trading

Engineer who ships AI-powered SaaS to paying users.Multi-tenant. Multi-agent.Finance & trading workflows. Spec, code, deploy, monitor — same hand.

Four years building AI-powered SaaS end-to-end. Two flagship products live: stox.my and TradeClaw. Multi-tenant subscription, vision-LLM signals, RAG agents, multi-agent persona systems. Real users, sustained ops, p50 80ms on production read paths.

2 flagship SaaS
Live · paying users
Multi-tenant
Subscription · auth · billing
Multi-agent
Personas · memory · modalities

Kuala Lumpur, Malaysia

AI-Powered SaaSMulti-TenantMulti-Agent PersonasVision-LLMRAG · bge-m3Real-Time Market DataSub-100ms ReadsLive Production Users0-to-1AI-Powered SaaSMulti-TenantMulti-Agent PersonasVision-LLMRAG · bge-m3Real-Time Market DataSub-100ms ReadsLive Production Users0-to-1

By the numbers

What four years at the intersection of finance and AI actually looks like.

What I do

Three things, deeply. Everything else is a side project.

01

Trading platforms & investor analytics

Stock analytics SaaS, multi-factor screeners, watchlists, AI-assisted alpha discovery. Real-time pricing pipelines, hand-written SQL, sub-100ms reads on hot queries.

  • Real-time market data ingestion + ranking
  • Momentum + fundamentals factor screening
  • Investor-facing dashboards tuned for data density

02

Community platforms & multi-agent bots

Community apps with points, leaderboards, and social loops — like the Gold Traders portal. Contest platforms that pull live standings from sources with no public API. And a different kind of chatbot: dynamic, multimodal agents that talk to each other while humans watch and step in.

  • Community apps — points, leaderboards, social automation
  • Contest engines with API-less data scraping
  • Multi-agent chat — bots converse, humans observe

03

End-to-end ownership

Spec, design, code, deploy, monitor, iterate. Smallest useful MVP, validate with real users, then scale. Highest-leverage work first. Founder posture — even inside a bank.

  • Full TypeScript across the surface
  • Python / FastAPI on AWS (ECS, Lambda, RDS)
  • Cloudflare-native edge: Workers, D1, KV, R2, Durable Objects

Selected finance work

Finance systems built for real operations.

Trading SaaS, investor analytics, trader engagement and production AI. Finance outcomes and operating context stay visible first.

Flagship · live SaaSCase study

TradeClaw

Solo · backend + frontend + vision-LLM · live subscription product

AI-powered trading signals SaaS. FastAPI + React, multi-tenant subscription product, vision-LLM chart parsing across multiple asset classes. Live paying users. The full shape of a modern AI SaaS — auth, billing surfaces, agent inference, the lot.

AI SaaSVision-LLMMulti-tenantFastAPIReact

Flagship · live SaaSCase study

stox.my

Solo · founding engineer · 0-to-1 to paying users

Stock analytics & AI alpha-discovery SaaS for Malaysian markets. Real-time Bursa pricing, multi-factor screener, watchlists, AI-assisted alpha discovery. Edge-first, sub-100ms reads on hot queries. Founding engineer, 0-to-1 to paying users.

AI SaaSMulti-factor screenerEdgeBursa Malaysia

ProductionCase study

GoldTraders Trader Portal

Solo · 1 year · replaced what would have been a 3-engineer + PM build

Trader-facing portal with AI signals, leaderboards, community automation. React 19 + Hono, 15+ REST endpoints, Durable Objects rate limiting, 4-mode contest engine with tamper-evident credit-integrity audit trail. Lifted client trading-desk monthly net ~$50K → $200K within 1 year.

React 19HonoAI signalsAudit trail

Production AI agent

RoboForex Support Bot

Solo · prompt + context + retrieval engineering end-to-end · in production

Production AI agent — RAG with bge-m3 embeddings, query rewriting, multi-tenant routing, follow-up persistence. Python / FastAPI on AWS. Prompt + context + retrieval engineering owned end-to-end.

RAGbge-m3FastAPI on AWSMulti-tenant

More production work

5 systems, kept compact.

Professional engineering proof without another media gallery.

Beyond finance4 public products beyond financeExplore 4 products
01Live

Reflection companion

Qur'an Tadabbur

Choose a verse. Slow down. Reflect with structure.

A calm, AI-assisted companion for personal Qur'an reflection, with guided entry points, thematic studies and saved reflections.

Visit sitequrantadabbur.com

Product notes

Experience
Select a verse, follow a guided reflection, then save useful thoughts for a later return.
Trust boundary
The product is positioned for personal study, not as a replacement for tafsir or scholarly guidance.
  • AI-assisted reflection
  • Guided study
  • Saved reflections
02Live

Mosque marketplace

MosRev

Find a mosque service without chasing separate contacts.

A Malay-first marketplace for mosque facilities, classes, community programs and vendors across Selangor and Kuala Lumpur.

Visit sitemosrev.com

Product notes

Experience
Search leads the journey, followed by clear categories for spaces, learning, community programs and services.
Trust design
Verification badges, payment cues and receipt language stay visible near the booking decision.
  • Malay-first
  • Service discovery
  • Booking trust
03Preview

Source-grounded learning

Dr. Saeed Foudah Study

Ask a question, then follow the source back to the book.

A bilingual study preview for Sheikh Dr. Sa'id Foudah's published works, organised around Ask, Learning Path and Library.

Visit sitedrsaeedfoudah.com

Product notes

Experience
One learning loop connects a source-grounded answer, a structured reading path and the underlying library.
Rights boundary
Full-book downloads stay unavailable until written permission and applicable rightsholder permission are documented.
Reliability
Generated answers are presented with an explicit instruction to verify them against the original works.
  • Bilingual
  • Source-linked
  • Rights-aware
04Live

Clinical communication practice

BedsideLoop

Practise the conversation before the exam room.

Voice-first ten-minute communication and ethics stations for MMed and PACES candidates, followed by rubric-based examiner feedback.

Visit sitebedsideloop.com

Product notes

Experience
Choose a station, speak with a simulated Malaysian patient, then review a structured examiner report.
Safety boundary
Patients, faces and scenarios are synthetic. Feedback is formative, and human examiners remain authoritative.
  • Voice-first
  • MMed and PACES
  • Formative feedback

Production evidence

Open the receipts when you need them.

Live product captures and role-scoped portals stay out of the way until you choose to inspect them.

Career line

Investment desk, then engineering, now both at once.

  • 2024 — now

    Full Stack Engineer & AI Trading Systems · RM Investment Bank Ltd

    Built a trader portal (apps, AI signal tools, leaderboards, community automation) that lifted average monthly net ~$50K → $200K within 1 year — replaced what would have been a ~3-engineer + PM build. 15+ production apps across investor dashboards, signal bots, RAG agents, weekly contest engines.

  • 2023 — 2024

    Founder / Technical Lead · Tibyan AI Sdn Bhd

    Founded an AI-first venture. Secured seed investment from Cradle and Nexea. Owned product, engineering, and GTM end-to-end across Python, TypeScript, React, PostgreSQL, AWS — first-engineer posture from day one.

  • 2021 — 2023

    Manager — Growth & Operations · Salmi Niaga Solution

    Drove ~400% increase in marginal profit through better systems, automation, and execution discipline. Built Python analytics dashboards and reframed P&L from gross-revenue chasing to unit economics.

  • 2020 — 2021

    Investment Analyst & Data Engineer · Inter-Pacific Asset Management

    Contributed to AUM growth from ~$10M to ~$200M in ~1.5 years through data-driven research and screening. Built data pipelines, factor models, and analytics dashboards in Python and SQL.

    Best Shariah Equity Malaysia 2020 — Lipper / Refinitiv

  • 2016 — 2020

    Finance Executive · AmanahRaya Trustees Berhad

    Fund accounting, treasury operations, financial modelling. Excel/VBA automation that compressed reporting cycles — first hands-on exposure to the data shape behind investor analytics.

Stack

The tools I actually use to ship — not a fashion show.

AI & LLM (primary)

  • Anthropic Claude
  • Claude Agent SDK
  • MCP (Model Context Protocol)
  • OpenAI
  • HuggingFace
  • Vision-LLM chart parsing
  • RAG · bge-m3 embeddings
  • Multi-agent persona runtimes
  • Prompt + context engineering

TypeScript & Node (primary, ~5y)

  • Node.js
  • React 19
  • Next.js
  • Vite
  • Hono
  • Express
  • REST APIs
  • Hand-written SQL

React Frontend (primary)

  • React 19 hooks
  • Suspense
  • Tailwind
  • Framer Motion
  • Data-dense dashboards

Python & Backend (primary)

  • FastAPI
  • Flask
  • Django
  • Pandas
  • scikit-learn
  • Factor models
  • ETL pipelines

Databases (primary)

  • PostgreSQL
  • MySQL
  • SQLite / D1
  • MongoDB
  • Redis-style KV
  • Numbered .sql migrations

AWS & Containers

  • EC2
  • ECS
  • Lambda
  • S3
  • RDS
  • IAM
  • Docker
  • GitHub Actions
  • Terraform

Edge & Distributed

  • Cloudflare Workers
  • Durable Objects
  • D1
  • KV
  • R2
  • Browser Rendering
  • Cron Triggers
  • Apache Kafka

Crypto / Onchain (breadth)

  • Smart contracts (Canton / Daml)
  • EVM signing · keccak256
  • L1 / L2 patterns

Contact

Have a problem at the intersection of finance, AI, and infra?

I read every message. Responses typically same-day on weekdays. If it is not a fit, I say so directly and point you to someone who is.

+60 11 6983 3882· phone / WhatsApp

For agency or NDA briefs, send a one-paragraph summary first.