I'm an aspiring 🚀 ML engineer and M.S. Computer Science student @ University of Illinois Chicago 🌆

I love building intelligent systems to solve real-world problems with cutting-edge technology. 💡

• Programming Mastery: Python (NumPy, Pandas, Scikit-learn, TensorFlow, PyTorch, JAX, HuggingFace, Keras), Java, C++, Scala, Shell scripting.
• Machine Learning Expertise: LLM Alignment, Deep Learning, Natural Language Processing, Computer Vision, Reinforcement Learning, Gradient Boosting.
• Data & Cloud Proficiency: Hadoop, Spark, MySQL, PostgreSQL, MongoDB, AWS (SageMaker, Lambda, RDS, EC2, ECS, S3, Amplify, Bedrock), GCP, Azure ML, Docker, Kubernetes, Terraform, Slurm.
• AI & Model Development: Distributed training (FSDP, DeepSpeed, Accelerate), Agentic workflows (LangChain, FastAPI), ML pipelines, deployment & monitoring.

• Founding Engineer – Quin (Mountain View, CA) (Jun 2025 – Present)

  • Architected backend Lambda + API Gateway routing for dev/prod environments ensuring scalability.
  • Integrated financial + LLM APIs for user insights.
  • Implemented a RAG agent with math tools + advanced search.
  • Designed AWS RDS + S3 pipelines for structured + open-source LLM data.
  • Tech Stack: Python, AWS (Lambda, RDS, ECS, S3, API Gateway, Amplify), TypeScript, Restful APIs.

• Graduate Research Assistant – University of Illinois at Chicago (Aug 2024 – May 2025)

  • Built distributed online-DPO training pipelines with LLM-as-a-Judge, cutting training time 54% on A40/A100 clusters.
  • Developed an RL agent (PRBC) for cancer treatment plans, achieving 98% simulated reward and 61% macro-F1 matching doctors’ choices.
  • Tech Stack: PyTorch, Kubernetes, Slurm, W&B, Accelerate, DeepSpeed.

• Graduate Research Assistant – UI Health, Neurology & Rehabilitation Dept. (Oct 2023 – May 2025)

  • Deployed an ML pipeline reducing EEG processing time by 40%.
  • Achieved 99% accuracy, 88% sensitivity, 0.43 FP/hr in seizure detection.
  • Tech Stack: Python, MySQL, JAX, SciPy, Docker, Azure ML Studio.

• Machine Learning Engineer – Cactus Communications (Jun 2022 – Jul 2023)

  • Cut GPT-3.5 inference costs by $10K annually.
  • Built summarization + keyword extraction tools for scientific text.
  • Optimized AWS Inferentia deployments for low latency + cost efficiency.
  • Tech Stack: Python, PyTorch, HuggingFace, AWS (EC2, S3, Inferentia, Lambda, CloudWatch), Docker, Terraform.

• Online LLM Alignment Beyond Preference Methods (Under Writing)
  Advanced LLM alignment via online DPO with LLM-as-a-Judge, reducing reliance on preference-based scoring with automated W&B hyperparameter sweeps.

• Guided Policy Gradient for Dynamic Treatment Plan Prediction with Symptom Burden in Cancer (Under Publication, Apr 2025)
  Developed and evaluated a Deep RL framework to automate cancer treatment planning, optimizing efficacy while minimizing symptom burden.

• Automated Seizure Detection in Ambulatory EEG (AESnet 2024)
  Built an EEG seizure detection system using XGBoost, CatBoost, and LightGBM ensembles, achieving 99% accuracy, 88% sensitivity, 0.43 FP/hr.

• Capacity Estimation of Li-Ion Battery Cells Using Deep Neural Networks (Springer, Oct 2024)
  Leveraged LSTM models for forecasting, achieving 0.0067 RMSE loss with statistical preprocessing.

• Exploring Large Language Models: Concepts, Alignment Techniques, and Practical Implementation (Medium, Dec 2024)
  Documented implementations of LoRA, QLoRA, SFT, RLHF, DPO, KTO, ORPO for optimizing model behavior.

• Notey – AI-Powered Memory Companion 📝
  Built a multi-modal event capture platform with audio transcription, AI summarization, semantic search, and secure Supabase storage.

  • Tech Stack: Python, FastAPI, React, TypeScript, Supabase, Whisper.cpp, Gemini AI, Vercel, Railway, Fly.io.

• LLM-Grounded Text-In-Image Generation 🖼️
  Leveraged LLaMA-16B to generate conditional text masks within images, fine-tuned GLIGEN with custom loss on A100 GPUs.

  • Tech Stack: Python, PyTorch, OpenCV, C++, Slurm, HuggingFace.

• Ambulatory EEG Signal Analysis 🧠
  Reduced EEG signal processing time by 40% and achieved 99% accuracy and 88% sensitivity in anomaly detection.

  • Tech Stack: Python, JAX, SciPy, MySQL, Docker, Azure ML Studio.

• Evaluating LLM Powered AI-Agents 🎮
  Developed agents for 7B/16B models to navigate RL-Gym games (blackjack, pathfinding), improving reasoning and reducing bias.

  • Tech Stack: LangChain, FastAPI, Python, PyTorch, HuggingFace.

• CS553 Distributed Computing Project – Distributed Algorithms Simulation 🖥️
  Simulated distributed algorithms with message-passing and shared memory, visualized with Prometheus and Grafana.

  • Tech Stack: Scala, Akka, Prometheus, Grafana, IntelliJ.

• SCIPASUMM 📜
  End-to-end research paper summarization pipeline using NLP models.

  • Tech Stack: Python, Bart-LS, HuggingFace, PyTorch, Transformers.

• Apple Grading Using Computer Vision 🍏
  Grading apples using image processing and CNNs to support agri-tech.

  • Tech Stack: Python, OpenCV, TensorFlow, Keras, CNN.

• Gaze Detection Using Computer Vision 👀
  Real-time driver gaze + drowsiness detection.

  • Tech Stack: Python, OpenCV, MediaPipe, TensorFlow Lite.

• UNET Implementation for Image Segmentation 🌟
  Exploring UNET architecture for segmentation tasks.

  • Tech Stack: Python, Keras, TensorFlow, UNet.

• YOLO Implementation for Mars Anomaly Detection 🚀
  Detecting Mars anomalies with YOLOv3.

  • Tech Stack: Python, YOLOv3, Keras, OpenCV, PyTorch.

• Tweet Topic Modelling 📊
  Topic modeling and visualization for tweets.

  • Tech Stack: Python, Gensim, NLTK, Matplotlib, Seaborn.

• Emotion Recognition with Face Mask 😷
  CNN to classify emotions from masked faces.

  • Tech Stack: Python, OpenCV, TensorFlow, Keras, CNN.

• Lithium-Ion Battery Forecasting 🔋
  Estimated battery capacity with ≤4% error using time-series models.

  • Tech Stack: Python, TensorFlow, Pandas, NumPy, LSTM.

• Robotic Arm 🤖
  Trained a simulated robotic arm with RL algorithms to grab objects.

  • Tech Stack: Python, PyTorch, RL-Gym, OpenAI Gym.

• Plant Disease Classification 🌱
  Image classifier identifying 38 plant diseases with ResNet.

  • Tech Stack: Python, TensorFlow, Keras, ResNet, OpenCV.

There's so much more I'm learning and building as an aspiring ML engineer.
Let's connect on LinkedIn! I'm always happy to network with others who are passionate about AI. 😊