I’m Oussama Zeddam, an AI and full-stack engineer based in Algiers. I turn complex business requirements into working software: AI agents connected to real operations, SaaS platforms with a sound foundation, and automation that removes work from your team’s day. Professional software engineering since 2013. Direct senior ownership from the first architecture decision to production.
Hire a senior full-stack developer for frontend, backend, APIs and AI integration. My work spans SaaS development, AI agent development, RAG systems and scoped fintech software engineering. Explore the Expensify fintech engagement.
AI that works inside your business
Connect assistants to your approved documents, catalogue and operational tools. Start with one useful workflow—answer an enquiry, find the right product, prepare a document—and define how people review or take over when the system reaches its limits.
Full-stack software you can build on
Bring the product interface, APIs, permissions and data model into one coherent system. Whether you are launching a SaaS product or replacing fragmented internal tools, the architecture should make the next release easier to deliver.
An Algerian base. An international operating standard.
Work directly with an engineer in Algiers who also delivers for international teams. We can discuss the project in Arabic, French or English, establish a clear scope, and agree on milestones, documentation and the handover your team needs.
Viru store · London Book Fair Bookstore of the Year 2022
Catalogue-grounded AI discovery for Estonia’s celebrated bookseller. My work connects reader intent with structured product retrieval and catalogue synchronization, turning open-ended searches into relevant book recommendations.
Full-stack systems engineering for an Estonian EdTech platform with over 5,000 articles and tests in its public catalogue. Coordinated permissions, learning content and operational workflows connect administrators, teachers, students and parents.
Strengthening the infrastructure behind production AI agents: fail-fast validation against ambiguous tool dispatch, collision-resistant MCP tool names and application-controlled UI streaming. Implemented, regression-tested and merged into OpenAI’s official Python and JavaScript SDKs.
Full-stack engineering inside a publicly traded fintech. Root-cause analysis, a focused production fix and regression testing in a large React, React Native and TypeScript application, delivered through a completed five-star Upwork contract.
$2.3B in FY2025 sponsored research support, including SLAC
Selected production software delivery for Stanford University. Structured sentiment analysis, target extraction and evidence-grounded outputs, with abstention rules that keep unsupported inferences out of the decision path. The public technical note explains the architecture.
Nous Research · Repository stars observed November 2026
Merged upstream engineering in Nous Research’s widely adopted agent framework. Deterministic identity routing, alias resolution and per-user cache isolation strengthen multi-user sessions, with setup tooling for operators.
MCP-ready · Public APIs · Client production platform
Architected and delivered a live AI marketplace for a client: searchable skills, prompts and workflows, inspectable provenance, permissions and compatibility, plus public APIs and authentication. A full-stack product built for people and agent clients.
Architected and delivered a client platform that brings multi-model AI agents into business operations. Orchestration, RAG, knowledge connectors, BYOK and guardrails form one system for deploying, supervising and integrating specialized agents.
Explore the project
How we work
Close to the context. Accountable for the engineering.
Work directly with the engineer responsible for delivery
Agree on scope, milestones and acceptance criteria
Leave with working software and a usable handover
A good AI project starts with the operational constraint: missed enquiries, disconnected tools, slow access to information or a product that is difficult to evolve. I map the workflow, inspect the systems involved and identify a first release with a concrete acceptance criterion.
My connection to Algeria goes beyond an address. I founded DZ Combinator and the Algeria Freelancers Community, alongside a professional practice serving international companies. That combination keeps the business conversation grounded while the technical standard travels across borders.
For an Algerian company expanding internationally—or an international company looking for an engineer in Algeria—the engagement is built around the same fundamentals: named responsibilities, reviewable work, reliable communication and software your team can maintain.
Before the first conversation
Questions worth asking.
Who is Oussama Zeddam?
Oussama Zeddam is an AI and full-stack engineer based in Algiers, Algeria, working under the MAV brand. He has worked professionally in software engineering since 2013, holds a master’s degree in Software Engineering from Arizona State University, and was voted Exceptional Software Engineer at The Connected Awards, North Africa edition, in 2022.
How do I choose the best AI engineer in Algeria for my project?
Compare relevant production work, the engineer’s exact contribution, public code or client references, and their approach to testing, data access and handover. Review my case studies, merged OpenAI Agents SDK contributions and award coverage, then discuss the specific outcome your business needs.
Can you build the whole product as well as the AI?
Yes. My scope can cover the product interface, backend, APIs, database, authentication and AI workflows. We agree on the boundaries before work begins so the AI capability and the software around it can operate as one product.
Do you only work with clients in Algeria?
No. Algiers is my base, and my practice is international. I work remotely with teams in North Africa, elsewhere in Africa and worldwide. Location, communication windows and engagement terms are agreed around the project.
Your next production system
Bring the hard problem. Let’s build the way through.
Tell me what needs to work, what is blocking it and what success looks like. I’ll review the fit and help define the next move.