Afraz Khan

AI Engineer

NextGen Professional • Full-time

Jun 2025 - Present • 1 yr

AI Engineer — AI Agents, RAG & Document AI Ship production-ready AI agents, multi-agent systems and RAG chatbots that automate real backend processes — document review, support, lead handling and internal ops. Build retrieval-augmented generation assistants over PDFs, docs and databases using hybrid GraphRAG (Neo4j) + vector search (Pinecone, FAISS, Weaviate) with strict guardrails. Tech: LangChain, LangGraph, CrewAI, OpenAI, Claude, OCR, Document AI, computer vision, Python, FastAPI, Docker, AWS, GCP. Fast delivery, clean documented code, scalable architecture.

AI Engineer

Upwork • Part-time

Jul 2023 - Present • 2 yrs 11 mos

Here are three keyword-dense experience entries in paragraph form. Note Fiverr's experience description field also caps at ~600 characters per entry, so each is written to fit. AI Engineer — Fintech KYC & Identity Verification Built a production KYC/identity-verification pipeline that saved the client €40K/year. Engineered end-to-end OCR & Document AI for ID, MRZ and form extraction (PaddleOCR, AWS Textract, LayoutLM, YOLO) at 99%+ accuracy, plus ArcFace biometric face matching, liveness detection and a real-time transaction risk engine. Orchestrated verification agents with LangChain/LangGraph and deployed FastAPI microservices with Redis caching for sub-200ms latency. Stack: Python, Docker, AWS, OpenCV, computer vision, identity verification. AI Engineer — Healthcare AI & EMR Automation Developed a HIPAA-compliant AI medical scribe inside a live EMR, cutting clinical documentation time 60%. Built Whisper speech-to-text with speaker diarization, auto-generated SOAP notes and ICD-10 coding, wound-analysis computer vision, and clinical RAG with PII redaction and anti-hallucination guardrails. Integrated LLMs (GPT-4o, Claude) over patient records via vector + GraphRAG retrieval. Stack: Python, FastAPI, PyTorch, Hugging Face, NLP, healthcare AI, EMR automation, medical NLP. Freelance AI Engineer — AI Agents, RAG & Document AI Ship production-ready AI agents, multi-agent systems and RAG chatbots that automate real backend processes — document review, support, lead handling and internal ops. Build retrieval-augmented generation assistants over PDFs, docs and databases using hybrid GraphRAG (Neo4j) + vector search (Pinecone, FAISS, Weaviate) with strict guardrails. Tech: LangChain, LangGraph, CrewAI, OpenAI, Claude, OCR, Document AI, computer vision, Python, FastAPI, Docker, AWS, GCP. Fast delivery, clean documented code, scalable architecture.

AI Engineer

ACE Money Transfer • Full-time

May 2024 - Feb 2025 • 9 mos

At Ace Money Transfer, I designed and implemented machine learning models that provided actionable insights for critical financial operations. My responsibilities included architecting robust data preprocessing pipelines capable of handling large-scale and complex financial transaction datasets. I played a key role in driving cross-functional collaboration to ensure that AI solutions were aligned with the company’s strategic business goals. I also focused on optimizing model performance through rigorous testing and continuous refinement. Additionally, I contributed to the development of AI-driven systems that significantly enhanced operational efficiency and data-driven decision-making across the organization.