Ali Salloum

AI-Accelerated Full-Stack Architect — scalability, security, and production-readiness.

Blueprint diagram: reverse proxy to Manager, then routing to Unix sockets for supervisor and agents
25 → 1
Services to Manage
< 5ms
Proxy Overhead
CI/CD • Case Study 01

From Per-Agent Services to One Manager

I built the first version of our agent hosting: one service per partner agent, each with its own process and proxy config. It worked, but at 25 agents every release meant touching 25 services. I redesigned it into one manager per environment that provisions agents from the admin UI and routes requests to them over Unix sockets, so releases, backups and multi-server rollouts became predictable.

  • Replaced repetitive manual deployment steps with a consistent, operator-friendly workflow.
  • Made backups and multi-server rollout simpler and less error-prone through standardized service layout.
Backend • Case Study 02

Admin Campaign Orchestration on PocketBase

I built a notification orchestration subsystem on top of the app's PocketBase backend so non-technical admins can launch targeted campaigns without engineering help. The system handles audience selection, queued delivery through Firebase, real-time progress tracking, and full delivery telemetry so teams can see exactly what was sent, what failed, and what needs retry.

Technical focus

  • Designed a deterministic queue worker with per-device status transitions, retries, and idempotent processing to prevent duplicate sends.
  • Added an admin-facing control and observability layer: campaign states, delivery logs, completion guarantees, and failure diagnostics.
Web Scraping / Automation • Case Study 03

Partner Onboarding Data Engine for AI Q&A

I built the first version of this ingestion engine to speed up partner onboarding: it crawls a partner's domain, normalizes products and content into structured records, and loads them straight into the AI Q&A system. The team later extended it to single-page storefronts, and a five-site quality audit drove fixes for duplicates and missing fields.

Business impact
Faster partner onboarding and time-to-value
Structured data pipeline powering production AI Q&A
5
Partner sites quality-audited
AXIOMA-branded v2 assistant popup: greeting, example questions and a filled-in question ready to send
Live widget demo
Frontend • Case Study 04

LuukAI — Embeddable AI Assistant & Partner Widget

Partners drop in a conversational assistant: plain-language Q&A, smart recommendations, optional promos, and guided flows when needed. One codebase ships as a full web app and a floating embed—on-brand per tenant, multilingual, and built so the host site’s styling can’t break the experience.

Built with
Svelte (v2) and React (v1) · TypeScript · Vite — one pipeline, multiple surfaces (app, widget, embedded shell)

Highlights

  • The widget is visually self-contained—partner pages don’t accidentally restyle or break the assistant.
  • Surveys, partner add-ons, and analytics hook in cleanly so journeys feel native, not bolted on.
  • Per-brand copy, languages (including RTL), and tailored result screens—down to industry flows and booking handoff to the host.
REACT 18MUI V5VITE 5
Scholar impact

15citations on Google Scholar

Paper (ScienceDirect)Citing articles (Google Scholar)Recent citation: npj Computational Materials (Nature Portfolio) — materials discovery using quantum-annealed ML
Publication • Case Study 05

Quantum annealing in machine learning: Qboost on D-Wave quantum annealer

Peer-reviewed paper on quantum annealing and machine learning that extends my BSc thesis (I'm the second author, after Hadi Salloum). A QBoost-style ensemble is mapped to a quadratic unconstrained binary optimization (QUBO) formulation and run on a D-Wave quantum annealer. It explores when quantum hardware can complement classical ML pipelines, how experiments are set up on real hardware, and how to read results within device constraints.

Venue
Procedia Computer Science (Elsevier), 2024

Highlights

  • Connects quantum annealing with ML: QBoost framing executed on D-Wave hardware with a clear experimental narrative.
  • Sits at the intersection of QC and ML—relevant as both fields converge in research and applied work.
  • Growing citation footprint, including a 2026 citation in npj Computational Materials (Nature Portfolio).
Week of electricity price data: truth (blue line) vs model prediction (orange points), Dec 24–29, 2023
Full report (PDF)
ML / Forecasting • Case Study 06

Decision-Grade Electricity Price Forecasting

I built this forecasting system as a practical decision-support tool: it turns historical market signals into forward-looking price estimates that can inform planning, procurement timing, and risk controls. The pipeline covers data conditioning, supervised model training, and rigorous holdout evaluation, then exposes forecast behavior clearly so teams can act on trend direction and peak windows.

Operational takeaway
The model captures recurring intraday structure and turning points with useful accuracy for planning and timing decisions.

Highlights

  • Production-oriented evaluation slice: actual vs forecast on the same dense timeline for fast operator interpretation.
  • Captures strong cyclical price behavior and timing shifts instead of relying on naive baselines.
  • Evaluation outputs are decision-ready and easy to operationalize in planning workflows.
AI • Case Study 07

Agent Coach — long-memory Telegram assistant

A production Telegram bot where memory is a first class feature: goals and facts accumulate in simple files, load into every new chat, and improve after each reply. One codebase powers multiple specialist bots—same polished command surface, different expertise. Under the hood: Python with one-tap quality presets, optional live web lookup, and a backup model when traffic spikes so the chat keeps moving.

Model presets
Lean · balanced · max quality

Highlights

  • Session threads reset; long-term context does not—so coaching stays personal across weeks.
  • Shipped examples include fitness coaching and German exam prep; swap persona per deployment.

Core Architecture Philosophy

Performance-First

P99 latency is a requirement. Compute is optimized for cost and speed.

Security by Design

Security in schema, network topology, and CI/CD — not bolted on at the end.

Infinite Scalability

Stateless services and distributed state from ten users to ten million.

Ready to scale your vision?

Let's discuss production-ready architecture for your performance and security needs.

Schedule a ConsultationContact