Get in touch
Fouad Ikram · Product & Full-Stack Engineer

Software products, from the first conversation to production.

Get in touch Resume GitHub

Expertise

Most projects combine several of these.

01

Product engineering

Every project starts with the people who will use it. Requirement meetings with clients and stakeholders, finding where the real time is lost, scoping what to build first, and turning that into a system design. After launch, the same conversations continue: feedback, priorities and the next release.

  • Requirement meetings and scoping with clients
  • Workflow and interface changes based on how teams actually work
  • Clear trade-offs, estimates and release plans
02

Full-stack products

The product people sign in to and use every day: interface, API, database, accounts and payments, built so a team can keep extending it.

  • SaaS products and dashboards
  • Customer portals
  • Sign-in, payments and analytics
03

Assistants and agents

Assistants and agents that work from a company’s own documents, databases and spreadsheets, and show where each answer came from.

  • Support assistants for customers
  • Internal assistants for staff
  • Agents that handle bookings and follow-ups
04

Data pipelines

Pipelines that take files, PDFs, images, APIs and web pages in whatever state they arrive and turn them into tables a team can rely on.

  • Many sources merged into one database
  • Reports that refresh on their own
  • Market and pricing data collected on a schedule
05

Automation

Multi-step jobs that run without anyone pressing buttons. Each step is checked, failed steps retry, and anything unclear goes to a person.

  • Lead research and outreach
  • Content made and published on a schedule
  • Routine back-office work
06

Infrastructure and reliability

What keeps a product running after launch: cloud and server setup, security, monitoring, and systems that hold up when traffic spikes or an outside service fails.

  • Queues, worker pools and rate limits for heavy traffic
  • Retries and spending caps around AI providers
  • Cloud and self-hosted servers, secured and monitored

Experience

TechQuest.ai, 2024 to present. From engineering intern to leading a team of three. Client names are withheld.

AI & Full-Stack Software Engineer

TechQuest.ai · Jan 2025 to present

AI inventory intelligence platform

Full-stack RAG

Lets teams ask questions about their company’s private inventory data in plain English.

  • Answers come with live charts and analytical reports built right into the AI’s reply.
  • Flags at-risk stock early, such as items running low or items with safety concerns.
  • Sole designer and developer.

Multimodal AI platform for billboard advertising

Vision AI · Azure

Organises the spreadsheets, PDFs, links and thousands of photos that clients upload into structured records.

  • Vision AI matches each photo to its exact billboard and rejects photos that aren’t billboards, replacing manual review.
  • Deployed on Azure with Function Apps, a virtual machine, PostgreSQL and Key Vault.

Location-based AI research and video engine

Led a team of 3

Researches an area across Google, YouTube and Reddit, then turns what it finds into published videos.

  • Worked with the client in requirement meetings to shape the end product, then turned their needs into the system design.
  • AI writes the scripts; the system produces avatar videos with lip-sync, renders them in parallel with automatic retries and posts them to social platforms.
  • Led the build from architecture to production, self-hosted on a fully configured VPS (server, domain, Nginx).

Multi-channel ad campaign tracking platform

Full-stack · Azure

Brings spend, impressions, locations and creatives from 5 ad platforms into one view, used daily across billboards, mobile and TV.

  • Processes millions of impressions a day.
  • Worked directly with the client’s team to pinpoint slow processes and reshape workflows and the UI around their needs.
  • Secured with Defender for Cloud, firewalls, Key Vault, logging and alerting. Passed an independent security audit.

AI video production plugin for Claude

Voice AI

Makes polished product walkthrough videos automatically, the kind normally recorded by hand in Loom.

  • Writes a script for approval, plans what to show on screen, then records and edits the video.
  • Narration from ElevenLabs voice AI is synced word by word to what happens on screen.

AI Engineering Intern

TechQuest.ai · Jun 2024 to Aug 2024

Autonomous AI moderator for Discord

Does the full job of a moderation team, automatically and around the clock.

  • Handles warnings, kicks, bans and member management in real time, enforcing the community’s rules.

AI sales outreach platform

Writes personalized outreach emails in a company’s own voice.

  • Learns the business from its website, documents and links, including what it sells and its brand colors, tone and style.

Projects

Products built independently, from idea to launch. Each has a short demo video.

Watch demo · 1:30

Spelunk

AI agent

An autonomous AI agent that explores an unfamiliar codebase and answers questions about it, citing the exact files and lines it used. A verification ledger checks every citation against what it actually read, so it cannot invent sources.

Python · LLM agents · Tool useGitHub ↗
Watch demo · 1:37

Seatbelt

Infrastructure

A Go gateway in front of the OpenAI and Anthropic APIs with per-provider worker pools, a hand-written token-bucket rate limiter, retries with backoff, overload protection, and a hard real-time spending cap.

Go · ConcurrencyGitHub ↗
Watch demo · 1:51

Propla

Launched SaaS

An AI SaaS, designed, built and launched independently. It turns a short brief into a full client proposal, sent as a web page with e-signing, payments and analytics on which sections the client read.

Next.js · TypeScript · PostgreSQL · StripeGitHub ↗
· demo
Fouad Ikram sitting on the floor against a wall in a dark room, working on a MacBook, lit only by the screen

About

Hi, I’m Fouad.

I studied computer science at FAST-NUCES and have been building software professionally since 2024. The part of the job I like most comes after the demo works, when real data and real users start finding the problems.

These days I lead a small team. I sit with clients to work out what they actually need, design the architecture, write a good share of the code myself, and review my team’s code.

I also build my own products on the side. A few of them are up in Projects.

I like hard problems, boring infrastructure, and software that still works on a Monday morning.

EducationBS Computer Science, FAST-NUCES, 2021 to 2025
LanguagesEnglish (full professional), Urdu (native), Arabic (elementary)

Tools

AI & ML

LLM APIs, LangChain, RAG, AI agents, Tool calling, Embeddings, Vector search, Computer vision, Voice AI

Frontend

React, Next.js, TypeScript, JavaScript, Tailwind CSS

Backend

FastAPI, Flask, Node.js, Express, Go, REST APIs, GraphQL, Celery, SQL

Data & Cloud

PostgreSQL, pgvector, ChromaDB, MongoDB, Redis, Azure, AWS, Docker, Nginx, Linux VPS, CI/CD, GitHub Actions

Let’s talk.

Email is the quickest way to reach me.

Back to top
© 2026 Fouad Ikram