Aqsa Fayaz

B.S. Data Science graduate with hands-on experience building LLM-powered agent systems, multi-agent orchestration pipelines, and production chatbot backends. Co-Founder at Aangan, owning the full AI/ML workstream from LLM integration and tool-use APIs to production deployment.

Built real agentic systems using LangGraph and multi-agent coordination (InvoiceHound, Agentic GTBench), applied LoRA-based LLM fine-tuning and machine unlearning (SSU), and delivered NLP-driven document intelligence with DeBERTa and Gemini (BizRefine FYP). Developed a breast cancer awareness mobile app (Bibi) in partnership with Mobilink Bank and Shifa Hospital. Experienced in REST API development, Python backends, and rapid prototyping with LLM frameworks.

Targeting AI engineering roles focused on LLM agent development, RAG pipelines, and production AI systems.

Co-Founder & AI/ML Lead at Aangan (2025-02 – Present)

• Architected the full AI/ML workstream for an NLP-driven women's health app — data strategy, LLM integration, and production deployment
• Built and deployed an LLM-powered conversational agent with context-aware responses, tool-use integration, and REST API backend (Flask)
• Applied prompt engineering and context-aware retrieval to deliver culturally sensitive and medically accurate conversational responses
• Designed preprocessing pipelines, applied feature engineering and cross-validation to continuously improve model accuracy
• Led cross-functional collaboration between product, design, and engineering teams to translate requirements into ML deliverables

B.S. in Data Science – National University of Computer and Emerging Sciences