I’m a Computer Science B.S. + M.S. student at UC Riverside, expected to graduate in December 2026, interested in Machine Learning, Deep Learning, AI Agents, Generative AI, building thoughtful software, and engineering for good.

My work touches upon the intersection of AI, thoughtful software design + engineering, and accessibility; I particularly love turning my ideas into fun and impactful projects to benefit others. I'm an AI Trainer with Handshake AI and an AI product tester with DeepLearning.AI, where I focus on improving AI education through thoughtful course design and project experiences. I also spend my weekends teaching as an Islamic Studies Instructor at Institute of Knowledge (IOK), where I love cultivating a supportive and growth-driven learning environment for my students around Islam and what it means to live life as a Muslim in peace.

Here are a few AI courses I've directly contributed to and helped launch with DeepLearning.AI:

1. Multi-AI Agent Systems with crewAI
2. Long-Term Agentic Memory with LangGraph
3. Semantic Caching for AI Agents with Redis
4. Jupyter-AI Coding in Notebooks

• Aspiring AI/ML Engineer: I focus on building and learning with AI & ML (LLMs, Gen. AI, Multimodal AI, DL, RL, AI agents, etc.)
• Software Engineer: I enjoy building full-stack apps, intelligent tools, and friendly user experiences.
• Engineering for Good: My goal is to create projects that are technically impressive while being helpful and accessible to others.
• Beyond Code: I’m obsessed with boba, reading, skating, building LEGO, going for coffee runs, and gaming, which keep me curious and inspired.

• Snaption
  Explored multi-modal ML and built an end-to-end image captioning system with PyTorch to generate natural language descriptions of images. Trained on Flickr8k (8K images), tackling hyperparameter tuning, overfitting prevention, and data augmentation strategies.
  View my demo here and read about my project here.

• Text2SQL LLaMA Analyst
  Fine-tuned LLaMA-2-7B to translate natural language descriptions into structured, SQL queries using parameter-efficient fine-tuning techniques: LoRA + 4-bit quantization.
  Includes a zero-setup Colab Demo

• GPT From Scratch
  Implemented multi-head self-attention, transformer blocks, and autoregressive decoding entirely from scratch in PyTorch to truly understand how GPT models work.
  Test it out yourself here.

• Pixelate
  Developed a colorblind-friendly pixel-art editor with an AI assistant (Pixi), real-time colorblind-friendly filters, and an accessible, intuitive interface for artists.
  View our demo here.

• Reddit Crawler
  Engineered a robust Reddit crawler and search engine that scraped, cleaned, and enriched 500MB+ of real-time subreddit data with multi-threading support, improving crawl throughput by 4× via parallelization.
  View our demo here.

• Deep Learning & Agentic AI
• Reinforcement Learning
• LangChain & LangGraph Foundations
• Multi-Modal and Multi-Agent ML Systems
• AI Engineering Applications
• AI for Creative Workflows and Developer Productivity Tools

• Portfolio: aamir-khans-portfolio.vercel.app
• LinkedIn: linkedin.com/in/aamir-khan-aak521
• Learning: coursera.org/learner/aamir-khan-0365
• Resume: drive.google.com/file/d/1PrH3AKYS2jmBpvbnd091-TFjmjRBFDBX/view
• Email: akhan197@ucr.edu | aamirksfg@gmail.com

I'd love to use AI/ML and AI engineering to solve complex, interesting problems that will help us live more productive lives! My dream is to build AI tools that have the potential to help millions around the world and improve our quality of life as developers and learners.