AI Developer & Computer Science Student | Agentic Workflows & Systems Architecture
I build autonomous reasoning systems, high-precision retrieval engines, and developer tooling. My mission is elevating AI from conversational models to operational agents that can execute complex tasks with complete safety and program relevance.
- 🤖 Agentic Systems: Building autonomous task-loops using multi-agent frameworks (LangGraph) and state machines.
- 🧠 RL & Latent Alignment: Researching reasoning limits and Self-Distillation degradation in RL-based agents.
- 🔍 High-Precision RAG: Designing contextual indexing pipelines to enable semantic queries across codebases.
- 🛠️ System Engineering: Compiling type-safe schemas to establish zero-overhead Roblox execution environments.
"The future belongs to agents that act, not just models that predict. I engineer the logic that lets them execute safely."
| Project | Description | Stack | Status |
|---|---|---|---|
| 🤖 OpenEnv-CR | Code-review simulation environment for LLMs & RL agents | Python, Reinforcement Learning | ⚡ Active |
| 🧠 docChat | Interactive RAG application to chat with raw codebases | Python, LangChain, embeddings | ⚡ Stable |
| 🔗 Roblox Event Bridge | Code-generator for type-safe Luau schemas on Roblox | Python, Pydantic, Code Gen | ⚡ Active |
| 📈 DRSP-ERL Analysis | Mitigating reasoning collapse in self-distillation models | PyTorch, Latent Alignment | 📈 Research |
| 📚 Daily Notes | Knowledge base of daily software architectures, eBPF logs, and DSA | Markdown | 📂 Docs |
- ✅ Advanced LLM Agentic Workflows & Tool Architectures
- ✅ High-Precision RAG Chunking & Retrieval Strategies
- ✅ Differentiable Reflection-State Preservation (DRSP) in RL
- 🔄 Type-Safe Code Compilation & Parsing Paradigms
- 🔄 low-level eBPF & Kernel Instrumentation
Feel free to reach out if you want to collaborate on AI projects, discuss Agentic workflows, or just chat about technology!
