Abdelrahman Abouroumia (Romia)Backend and AI systems engineer.

1. Sep 2026 – Present

   

   Software Engineer, AI Systems

   Tutanic

   Backend services and AI systems.

   Microservices · FastAPI · Django · NATS JetStream · PostgreSQL · PgBouncer · Dragonfly · LangGraph · CopilotKit · MCP · AWS Bedrock · Amazon S3 · Kubernetes · Helm

2. May 2025 – Present

   

   Software Engineer & Co-Founder

   Zaylon AI

   Co-founded Zaylon and built its AI system and backend: a LangGraph tool-calling agent and tri-lingual NLP on FastAPI microservices that talk over Kafka, with PostgreSQL, Redis, multi-tenant isolation, and per-service Railway deployment, built for MENA merchants selling on WhatsApp, Instagram, and web.

   • Built the conversational AI in LangGraph as a single tool-calling agent with about 35 scoped tools across sales, support, and checkout. An earlier multi-agent supervisor was kept behind a flag and deleted once it was clear the single agent was the only path ever serving a request
   • Built hybrid product search combining keyword matching, semantic vector similarity, and Reciprocal Rank Fusion scoring
   • Wrote the FastAPI Python backend and its service architecture, containerized with Docker for deployment
   • Built multi-tenant security: argon2id key auth, per-merchant row-level isolation, and semgrep rules that fail CI on raw table access outside the repository layer
   • Built a Postgres-backed saga engine with reverse-order compensation so a failed checkout step rolls back cleanly
   • Batched rapid-fire WhatsApp messages in Redis into a single turn before invoking the agent, so it reads the whole thought at once instead of replying to each fragment

3. Sep 2024 – Feb 2026

   S

   AI Coding Expert

   Scale AI · Contract

   Reviewed and improved LLM-generated code for RLHF training data used to train frontier AI models.

   • Reviewed and ranked code repositories across languages for RLHF data pipelines used to train frontier AI models
   • Built test suites as quality control, scoring correctness, efficiency, style adherence, and edge case handling
   • Reviewed thousands of code samples, providing detailed feedback on reasoning chains, code structure, and technical accuracy

4. Jul 2025

   

   Software Integration Engineering Intern

   Ejada

   Worked on core banking integration systems, building APIs and event-driven messaging pipelines.

   • Built RESTful API endpoints for core banking operations using Spring Boot with query optimization and Redis caching for frequently accessed account data
   • Configured Apache Kafka producers and consumers for asynchronous transaction processing, decoupling payment events from the main request lifecycle
   • Implemented API gateway security with OAuth2 authentication, rate limiting, and request validation using WSO2 API Manager

5. Aug 2024 – Oct 2024

   

   Automotive Cybersecurity Intern

   Swift-ACT

   Contributed to automotive cybersecurity research, building security scenario catalogs and running attack simulations in vehicle simulation environments.

   • Built security scenario catalogs with threat models, attack vectors, and mitigations for CAN bus and ECU communication systems following automotive cybersecurity standards
   • Conducted fuzz testing on Automotive Emergency Braking (AEB) components, identifying edge cases in sensor fusion timing and brake actuation logic under corrupted input data

• Aug 6, 2026commentdraft: replying in public on a client's behalfPaid work: drafting public replies under a client's name, where a wrong price is a bill the client pays. The four ways the obvious build fails them, the approval gate that five pasted characters walked straight through, the three platforms whose comment text is not what the person typed, and a measured run where the bargain model billed 27 times more per comment.14 min read
• Jul 5, 2026Talos: a document-grounded team assistant, and the retrieval evaluation behind itA walkthrough of Talos, a team chat platform whose assistant answers from a team's own uploaded documents with citations. It covers the retrieval pipeline (hybrid search, cross-encoder reranking, cited generation) and the paired evaluation that found and fixed the chunking bug behind weak answers, raising judged correctness from 0.657 to 0.855 on the workspace's own corpus.9 min read
• Nov 11, 2025Measuring your own writing voice, one function word at a timeHow I built a tool that turns a few of your writing samples into a measurable fingerprint, scores how far a draft has drifted from it, and nudges it back, with honest numbers about what the signal can and cannot do.11 min read
• Nov 1, 2025When a speech-emotion model is really just recognizing the actorsA speech-emotion model can look impressive and still be mostly memorizing the 24 RAVDESS actors. Here is the leak, the speaker-independent pipeline that replaced it (frozen WavLM, layer weighting, attentive pooling, calibrated late fusion), and the honest speaker-independent result: 78.8%.14 min read

06

Get in touch

If you are building agents, backend infrastructure or AI products, or hiring for one, I would like to hear about it. Email reaches me fastest.

aabouroumia@gmail.com

• GitHub
• LinkedIn
• Hugging Face
• Résumé (PDF)