AI Engineer focused on AI Agents, Deep Learning, and NLP.
Self-taught, engineering strong end-to-end AI systems just for fun.

• AI Agent architectures & tool-using systems
• Deep Learning with PyTorch
• NLP model development & modern transformer techniques
• Reinforcement Learning algorithms

A custom deep learning optimizer inspired by biological learning and gradient-based updates.
Designed to experiment with alternative optimization behaviors beyond Adam/SGD.
Inspired by Karpathy's micrograd.

A reinforcement learning agent trained using Proximal Policy Optimization for algorithmic trading.
Focus on stable returns, reward engineering, environment design, and policy performance monitoring.

An AI-powered chess engine, combining:

• NNUE-based evaluation
• Search-based planning
• Agentic decision loops
  A personal research project to explore competitive game-playing AI.

• Python, Linux, Git/GitHub
• Docker, FastAPI
• Experiment tracking (Weights & Biases)
• Google Colab / Kaggle workflows
• Model deployment fundamentals

• PyTorch, NumPy, Pandas, Scikit-Learn, Matplotlib/Seaborn
• Transformer models, attention mechanisms
• CNNs, RNNs, sequence models
• Optimization algorithms (including custom ones like BitGrad)
• Reinforcement Learning (PPO, DQN fundamentals)

• Hugging Face transformers, SpaCy
• Tokenization, embeddings, sequence modeling
• LLM fine-tuning & prompt engineering
• Text classification, generation, and evaluation

• Advanced AI Agent frameworks & multi-agent systems
• Scaling laws, evals, and reliability for AI systems
• Production-grade NLP pipelines
• Chess engine search optimization (Echelon)

Open to collaboration, contributions, and interesting discussions in AI.
Let’s build something impactful. Email: [email protected]