You're seeing this page as if you were . The main menu is still yours, though. Exit from immersion
Malt Strategy
New report
Malt Tech Trends 2026
Abdulrahman ElsmmanyAE

Abdulrahman Elsmmany

Senior AI/Software Engineer

€551/day
Shibīn al Kawm, EG
3-7 years
artificial intelligence
AI Agent
Python
Typescript
Langchain

Average response time: 1 hour

About Abdulrahman

Production AI systems that actually ship, not notebook demos. I've spent five years building voice agents, multi-agent platforms, and RAG pipelines, and I own the full stack around them: Postgres schema, Cloud Run deploys, Next.js frontend, CI/CD.

Recent work:

Hams.AI: real-time voice AI on Pipecat + WebRTC (Daily), with streaming STT-LLM-TTS across OpenAI, Groq, ElevenLabs, Deepgram, AWS Transcribe/Polly. Multi-node conversational flow engine with conditional branching, variable extraction, and agent transfer. Concurrent production calls over Twilio and Telnyx.

PsHub: Next.js 16 / Supabase B2B marketplace, 68K+ LoC monorepo. LangGraph multi-agent system (Supervisor + Planner + 5 workers, 26+ tools) with risk-classified HITL approvals and streaming chat over OpenRouter. 64-table Postgres with 350+ RLS policies, pgmq async queuing, Google Cloud Run.

Addis Care: HIPAA-compliant 5-agent healthcare intake on Firebase Genkit + LiveKit, with RAG over resident notes for real-time ADL/IADL summaries to care staff.

Profit 717: crypto platform for 50,000+ users. BEP-20 smart contracts for automated token payouts, Python asyncio/aiogram backend, Redis-backed real-time profit calc.

What I usually handle:

  • Voice AI architecture (Pipecat, LiveKit, Daily, Twilio, Telnyx, WebRTC)
  • Multi-agent orchestration (LangGraph, LangChain, Firebase Genkit, Pydantic-AI)
  • RAG pipelines: pgvector, BM25 hybrid retrieval, cross-encoder reranking, semantic caching
  • The full production stack around AI: FastAPI, Next.js, Supabase, GCP, Kubernetes
If you're building a voice agent, a multi-agent platform, or any production AI system that needs more than a prompt, that's my lane. Drop me a line with what you're building and I'll tell you honestly if I'm the right fit.

  • English

    Native or bilingual

  • Arabic

    Native or bilingual

Remote only
Primarily works remotely

Experience

  • Addis Care
    Senior AI/ML Engineer & Full-Stack Architect
    TECH
    February 2025 - Today (1 year and 4 months)
    California City, United States
    • • Engineered HIPAA-compliant application architecture with secure authentication and data handling across all services and AI modules.
    • • Designed a multi-agent conversational AI system in Firebase Genkit with an orchestrator coordinating task delegation across specialist agents.
    • • Built a HIPAA-compliant clinical decision support agent using RAG over resident notes, delivering real-time ADL/IADL summaries to care staff.
    • • Developed a real-time LiveKit voice agent and a Gmail-integrated RAG system for natural voice and email-based queries, backed by a Next.js/React full-stack implementation.
    Next.js artificial intelligence AI Agent Firebase Google cloud
  • Hams.AI
    AI/Software Engineer
    TECH
    January 2026 - March 2026 (2 months)
    Riyadh, Saudi Arabia
    • • Architected production backend for a real-time voice AI platform with FastAPI, SQLAlchemy, PostgreSQL, Redis caching, and Alembic migrations.
    • • Built streaming STT-LLM-TTS voice servers on Pipecat + WebRTC (Daily) with multi-provider support across OpenAI, Groq, ElevenLabs, Deepgram, and AWS Transcribe/Polly.
    • • Designed a multi-node conversational flow engine with conditional branching, variable extraction, and agent transfer, validated through Pydantic 2.x schemas.
    • • Delivered multi-agent orchestration with squad-based handoff, integrated with Twilio, Telnyx, and Daily for concurrent production call handling.
    Python artificial intelligence voip LLM FastAPI
  • PSHub
    Full-Stack Architect & AI Engineer
    TECH
    July 2025 - Today (11 months)
    Saudi Arabia
    • • Architected a Next.js 16 / React 19 / Supabase B2B marketplace (68K+ LoC monorepo) connecting Gulf event organizers with suppliers.
    • • Engineered "Hubby," a LangGraph v1 multi-agent system (Supervisor + Planner + 5 workers, 26+ tools) with risk-classified HITL approvals and streaming chat over OpenRouter with model fallback.
    • • Implemented a 3-tier AI memory system combining BM25 + pgvector hybrid search, Redis semantic response caching, and auto-embedding pipelines for supplier-project matching.
    • • Designed a PostgreSQL schema with 64 tables, 350+ RLS policies, 45+ functions, 38+ triggers, 200+ indexes, and pgmq async queuing; deployed on Google Cloud Run with Cloud Build CI/CD.
    artificial intelligence AI Agent Langchain Next.js Supabase
Check out Abdulrahman's experience

Recommendations

Be the first to recommend Abdulrahman

Help this freelancer shine by sharing your experience working together.

These freelancer profiles also match your criteria

AgathaA

Agatha Frydrych

Backend Java Software Engineer

4.7

(3)

2

BaptisteB

Baptiste Duhen

Fullstack developer

4.6

(4)

5

AmedA

Amed Hamou

Senior Lead Developer

4

(2)

7

AudreyA

Audrey Champion

Web developer

4.3

(3)

4

Signup to reveal

Education

  • B.S. in Artificial Intelligence
    Menofia University
    2026
    B.S. in Artificial Intelligence

Skill set

Web development

Categories