Kategorien durchsuchen
Entdecken
Fiverr Pro
Deutsch
$
USD
Talha Razzaq
@talharazzaq194
AI Engineer, Voice AI and Workflow Automation Specialist
Pakistan
Englisch
Über mich
I'm an AI Engineer building agentic systems, voice AI agents, and workflow automations for businesses that want to cut manual work and respond to customers faster. My background spans LangChain/LangGraph agent architecture, Deepgram + LiveKit voice pipelines, and n8n/Zapier integrations — backed by a Computer Engineering degree and hands-on production experience. I build systems that actually run, not demos, and I explain what I've built in plain language so you always know what you're paying for
Kompetenzen
Talha Razzaq
offline •
Meine Dienstleistungen
KI-Implementierung und -Bereitstellung
I will build n8n automation, zapier workflow and api integration
KI-Implementierung und -Bereitstellung
I will build ai voice agent, ai receptionist with vapi, retell and livekit
Arbeitserfahrung
AI Engineer
Z360 • Vollzeit
Aug 2025 - Present • 1 yr 1 mo
Built LLM evaluation and observability infrastructure in Python using LangFuse and LangChain, instrumenting multi-stage agent pipelines to trace individual model calls, isolate failure modes, and compare model and prompt variants against measurable latency and reliability baselines. • Led open-source development of Clave (clave.z360.biz), an AI-native project management platform combining issue tracking, real-time collaborative docs, and autonomous agents; owned repository structure, containerized development environments, and code review across contributors. • Engineered agentic workflows with LangGraph and BrowserUse that autonomously plan and execute multi-step tasks with retry and escalation logic, cutting manual intervention by 30%. • Designed production RAG pipelines (ingestion, chunking, embedding, retrieval) with LangChain and FAISS, exposed through Python REST services and deployed as containerized workloads. • Built low-latency real-time voice agent pipelines (STT → LLM → TTS) with Deepgram streaming ASR and LiveKit WebRTC, tuning end-to-end latency and turn detection for production deployments.
