Prompt Engineering for Scientific Q&A
Developed optimized prompts to improve factual accuracy and consistency of answers in technical domains such as biology, physics, and machine learning.
Overview
This page highlights my work on optimizing large language models (LLMs), covering areas such as:
- Prompt Engineering Techniques
- Chain-of-Thought and Role-Based Prompting
- RLHF (Reinforcement Learning with Human Feedback)
- Supervised Fine-Tuning (SFT)
- Token Efficiency and Inference Control
- Ethical and Bias Mitigation Strategies
Featured Projects
Chatbot Persona Development
Created dynamic personas for customer service LLMs, enabling context-aware and emotionally intelligent interactions using role-based prompt strategies.
Human-in-the-Loop Feedback System
Implemented RLHF feedback loop for refining outputs of generative models using real user interactions and preference scoring.
Tools & Technologies
AI generated image. Tools and Technologies
- OpenAI GPT-4 / ChatGPT API
- Anthropic Claude
- Hugging Face Transformers
- LangChain and LlamaIndex
- Pandas, NumPy, Python
- Vector Databases like Pinecone, FAISS
Get in Touch
For collaboration or demo requests, feel free to contact me Contact or visit my homepage.
www.HoqueAi.com