# Projects

Source: https://www.alouaniyoussef.com/projects

- [Osoul platform](https://www.alouaniyoussef.com/projects/osoul.md) (platform): Revenue as a Service for fine dining. A multi-tenant SaaS that captures the data restaurants usually lose and turns it into revenue KPIs.
- [Saudi-dialect speech-to-text](https://www.alouaniyoussef.com/projects/saudi-stt.md) (ai): Whisper large-v3 fine-tuned for Saudi dialect, with an audio pipeline for WhatsApp and Asterisk calls, behind live transcripts and tool-using voice agents.
- [Customer Voice: RAG pipeline with an MCP retriever](https://www.alouaniyoussef.com/projects/customer-voice.md) (ai): A RAG pipeline over structured data, documents and customer interactions, with an MCP server as its main retriever and auth, privacy and retention built in.
- [archio: object storage in Go](https://www.alouaniyoussef.com/projects/archio.md) (backend): Object storage in Go with a multi-node, high-availability cluster and semantic search over stored files, built on Docling.
- [Osoul deep-dive: multi-tenant platform with owned provisioning](https://www.alouaniyoussef.com/projects/osoul-multi-tenancy.md) (platform): Custom onboarding and Osoul-owned tenant provisioning, with full isolation of data, identity and configuration between restaurant groups.
- [Osoul: an AI assistant that acts as the user](https://www.alouaniyoussef.com/projects/osoul-ai-assistant.md) (ai): A LangChain assistant in the Osoul Portal that redacts personal data, calls tools with the user's own permissions, and asks before it writes anything.
- [Osoul: recommendations and dynamic pricing](https://www.alouaniyoussef.com/projects/osoul-recommendations.md) (ai): Dish and beverage recommendations, pairings and meal-completion promos in the Guest App, plus event-based dynamic menu pricing, on one menu data model.
- [Osoul deep-dive: secure NFC/QR table sessions](https://www.alouaniyoussef.com/projects/osoul-table-sessions.md) (platform): How a tap on a table's NFC tag or QR code opens a short-lived session for an anonymous diner, bound to one table and one visit.
- [AI Stones: an AI harness for Claude Code](https://www.alouaniyoussef.com/projects/ai-stones.md) (ai): Markdown skills, guardrail hooks and JSON settings that protect and improve how Claude Code operates in a project, installed per project and safe to re-run.
- [Osoul: server-side tracking and CAPI sync](https://www.alouaniyoussef.com/projects/osoul-server-side-tracking.md) (backend): A tracking service built on the open-source tracking kit, turning guest events into signed server-side Conversions API events for ad-audience optimisation.
- [Sowtek: omnichannel chat and channel integrations](https://www.alouaniyoussef.com/projects/sowtek-omnichannel-chat.md) (backend): Backend services for Sowtek's omnichannel inbox, with WhatsApp, email, X, Instagram and Facebook integrated behind one conversation model.
- [Sowtek: VoIP, PBX processing and sentiment detection](https://www.alouaniyoussef.com/projects/sowtek-voip.md) (backend): Backend for Sowtek's call centre on Asterisk: PBX event processing, call recording pipelines and sentiment detection on transcribed calls.
- [Sowtek FlowBuilder: an in-house low-code engine](https://www.alouaniyoussef.com/projects/sowtek-flowbuilder.md) (platform): The engine behind FlowBuilder, Sowtek's low-code builder for call, chat and SMS workflows, from IVRs to database lookups and agent routing.
- [Finamaze: from B2C to multi-tenant B2B](https://www.alouaniyoussef.com/projects/finamaze-migration.md) (backend): Migrating Finamaze from a B2C robo-advisor app to a multi-tenant B2B platform for financial institutions, with the infrastructure adapted to match.
- [Finamaze: investor onboarding, KYC and reconciliation](https://www.alouaniyoussef.com/projects/finamaze-kyc.md) (backend): The investor onboarding flow with document and visual identity checks, and reconciliation that keeps the platform's ledger in step with the bank.
- [Public chatbots on Botpress, Rasa and an in-house engine](https://www.alouaniyoussef.com/projects/chatbots.md) (ai): Chatbots built on three engines, from the bot backend and the chat widgets to the security a bot needs when anyone on the internet can talk to it.
- [Qalqul: Botpress integration](https://www.alouaniyoussef.com/projects/qalqul-botpress.md) (ai): Integrating Botpress v12 into Qalqul's contact-centre suite, so bots, human agents and tickets share one conversation and hand-off path.
- [Qalqul: classical NLP/NLU engines](https://www.alouaniyoussef.com/projects/qalqul-nlu.md) (ai): Intent classification and entity extraction on language servers and Duckling, tuned per client, before LLMs made it the default.
- [Qalqul: stateless migration, real-time systems and task dispatching](https://www.alouaniyoussef.com/projects/qalqul-realtime.md) (platform): Moving Qalqul's services to stateless, horizontally scaled processes with real-time messaging, a task dispatcher, and the infrastructure behind them.
- [10xJobs: job ingestion pipeline](https://www.alouaniyoussef.com/projects/10xjobs-ingestion.md) (ai): Collection, preprocessing, type-safe LLM extraction and matching: the pipeline that turns raw job posts into matches for each user's published profile.
- [10xJobs: search optimisation with segmentation](https://www.alouaniyoussef.com/projects/10xjobs-search.md) (backend): Matching jobs to profiles like a hash table: both are reduced to segment keys, so each job is only compared with the profiles in its bucket instead of every profile.
