Open to opportunities

SamirAkramOUNIS

Full-Stack / AI Engineer — Algiers, Algeria

I build production web platforms, computer-vision pipelines, and LLM-integrated systems with TypeScript, Next.js, NestJS, FastAPI, and Python. I’m a Master’s student in Intelligent Computer Systems Engineering, focused on full-stack product engineering and applied AI.

Years Experience

Featured Projects

Case Studies

Awards

View Projects

Download Resume

Scroll

Featured Projects

Pinned projects recruiters should open first.

Five selected projects across production platforms, AI research, client delivery, and agentic planning.

Full archive

Product EngineeringInternal

TPMS Expertic

Built role-based modules for a TPE operations platform managing bank requests, tickets, inventory, attachments, Excel workflows, and notifications using Next.js, NestJS, Prisma, PostgreSQL, JWT, CASL, and SSE.

Next.jsTypeScriptReactTailwind CSSshadcn/ui

Why it matters

Shows production full-stack delivery on an internal operations system: scoped bank/account-manager workflows, protected APIs, inventory tracking, attachments, notifications, and Excel import/export.

Full-stack contributor during my Expertic role. I worked on role-based dashboards, API integration, ticket workflows, inventory modules, and sanitized portfolio evidence.

View details Case study Deployment evidence

AI & Computer VisionExperimental

PathOS / Prostate Lesion Detection

Research pipeline for prostate whole-slide images using OpenSlide preprocessing, HistoEncoder/Virchow2 embeddings, artifact gating, ABMIL/TransMIL, QWK evaluation, and attention heatmaps.

Python 3.12PyTorchCUDAOpenSlideOpenCV

Why it matters

Strong AI/research signal: gigapixel WSI handling, pathology foundation-model embeddings, artifact-aware MIL, ablation-style evaluation, and explainability outputs without claiming clinical readiness.

Research and ML engineering work on a prostate WSI pipeline. The repository is private; the portfolio shows architecture, experiment artifacts, and model-evaluation evidence. This is not a clinical product.

View details

AI & Computer VisionPrototype

Smart Aquaculture AI / Firmatech

Prototype smart aquaculture platform with fish counting, disease checks, biomass estimates, pond monitoring, alerts, reports, mobile flows, and an admin dashboard.

FastAPIPythonYOLOv8OpenCVRoboflow

Why it matters

Shows applied AI beyond a notebook: model endpoints, tracking logic, generated reports, simulated sensor data, notifications, and operator interfaces.

AI/Research/Testing contributor in the Junction Algiers hackathon team. I worked on AI modules, data analytics, testing, and system evidence from earlier smart-fish-farming research.

View details Case study GitHub organization

LLM & Agentic SystemsPrototype

SynexFlow AI / Tattwir Hackathon

Second-place Tattwir prototype for AI-assisted business planning: sales forecasts, production planning, inventory alerts, external context, and action recommendations.

FastAPIPythonTimeGPTNixtlaLangGraph

Why it matters

Shows agentic AI tied to business data: REST APIs, forecasting, LangGraph workflows, dashboard screens, and public repos from a winning hackathon team.

Hackathon team member in the second-place SynexFlow team at Tattwir, hosted by Micro Club at USTHB. I present the team-built AI planning architecture without claiming solo ownership.

View details Case study Live landing

Client DeliveryClient Preview

Chouhrati

Arabic-first client preview with a public site, private CMS/backoffice, multilingual NestJS backend, store/order flow, media handling, and Algerian localization.

Next.js 16React 19TypeScriptTailwind CSSnext-intl

Why it matters

Shows current client delivery with Next.js 16, NestJS, Prisma, PostgreSQL, protected admin APIs, localized content, commerce flows, and preview deployment.

Nightbyte client project under active development. Public preview and sanitized architecture are shown; private repositories stay unlisted.

View details Client preview

Process

How I Build

Research, implementation, deployment, and measurement in one focused engineering loop.

01 / 04

Understand the workflow

I start by mapping users, roles, permissions, and the real operational flow before choosing screens or technologies.

RequirementsRolesWorkflows

system-design.md

Start from the workflow, not the UI.

Map users, roles, permissions, and core entities.

Define what each role can create, read, update, or approve.

Output:

- data model

- API boundaries

- role-based flows

02 / 04

Design the system boundary

I define the data model, API modules, authentication rules, and integration points so the backend supports the product clearly.

Data modelAPI designAuth

backend-architecture.ts

Build the backend around clear modules.

Keep auth, validation, permissions, and business logic explicit.

Use typed APIs so frontend and backend stay aligned.

Common stack:

Next.js + NestJS + Prisma/PostgreSQL

FastAPI for AI services when needed

03 / 04

Build the product loop

I connect the frontend, backend, database, and AI services into working flows that can be tested end to end.

Next.jsNestJSFastAPI

ai-pipeline.py

Treat AI as part of the product workflow.

Prepare data, run inference, validate outputs, then expose results clearly.

Avoid black-box demos when the user needs traceability.

Examples:

- computer vision endpoints

- RAG / LLM assistant flows

- evaluation artifacts and reports

04 / 04

Ship with evidence

I document the result with screenshots, architecture diagrams, live previews, or sanitized case studies so the work can be reviewed safely.

DeliveryDocumentationCase studies

delivery.log

Ship in small, reviewable increments.

Test the critical path before polishing visuals.

Document what is production, prototype, internal, or research-only.

Evidence:

- screenshots

- architecture diagrams

- live previews or sanitized case studies

Experience

Career Timeline

Professional experience across full-stack delivery, computer vision, and applied AI.

Software Engineer

Expertic

May 2025 - Nov 2025

Built internal service-management software and cloud tooling for TPE operations.

›Worked on TPMS, an internal platform for bank requests, service tickets, TPE inventory, attachments, Excel workflows, and notifications

›Contributed to service-management architecture for ITM intervention, installation, replacement, and support operations

›Configured Djezzy Cloud infrastructure and internal tools including Mattermost, GitLab, and management services

›Implemented role-based modules for account managers and bank employees using Next.js, NestJS, Prisma, PostgreSQL, JWT, CASL, and SSE

Deep Learning Engineer & Frontend Developer

White Bay Limited

Jun 2024 - Mar 2025

Built computer-vision and product features for internal CRM/ERP and AI assistant work.

›Created a YOLOv8 quality-assurance system for defective-product detection

›Developed React and TypeScript CRM/ERP features for internal business workflows

›Integrated chatbot agent flows with saved conversation history and reusable interface states

›Supervised two interns on a computer-vision safety-detection project

Intern

White Bay Limited

Mar 2024 - Jun 2024

Built an LLM virtual-assistant prototype with text and voice interaction modes.

›Developed a RAG-based assistant using Llama 2 for context-grounded responses

›Designed the interface in Figma and implemented the frontend with React and Tailwind CSS

›Built Laravel backend logic for context retrieval and agent interactions

›Implemented text-to-text, voice-to-text, and voice-to-voice communication modes

Skills

Technical Stack

Technologies grouped by current depth, so the strongest production stack is clear at a glance.

Production / Strong

TypeScriptReactNext.jsNestJSPostgreSQLPrismaDockerTailwind CSSGit

Working Experience

FastAPIPythonMongoDBLaravelYOLOv8OpenCVLangChain/RAGLLM integrations

Academic / Exposure

Spring BootRustFlutterSTM32TensorFlow

Contact

Let's Connect

Open to opportunities, collaborations, and technical conversations about engineering and research.

Get in Touch

Prefer email for initial reach out

samirounis2@gmail.com

Algiers, Algeria

Elsewhere

Find me across the internet.

GitHub

@akramoo

samirakramounis