Languages & Frameworks
Python
PyTorch
TypeScript
LangChain
LangGraph
Hugging Face Transformers
LlamaIndex
Building AI agents & LLM systems
Hi, I'm Denis Craciun
AI Engineer
I build production-grade AI systems — agentic RAG, multi-agent orchestration, MCP tooling and fine-tuned open-source LLMs. Turning research into things that actually ship.
About
A bit about me
AI Engineer focused on shipping production-grade AI systems: agentic RAG, multi-agent orchestration, MCP-based tooling, and fine-tuned open-source LLMs. I work end-to-end across the stack — Python and PyTorch, LangChain/LangGraph and Hugging Face, vector databases, QLoRA fine-tuning, ETL pipelines, model evaluation and hallucination mitigation — turning research into practical solutions, from tourist chatbots to self-improving agents. Driven by continual learning at the intersection of AI, software engineering, and finance.
Off the clock: football enthusiast & Muay Thai practitioner.
Skills
AI engineering stack
LLMs & Agents
Agentic RAG
Agent Orchestration
ReAct
Plan-and-Execute
Supervisor / Multi-agent
Reflection
Routing & Prompt Chaining
MCP
Tool / Function Calling
Agent Skills
Sandbox Execution
Persistent Memory
Goal Decomposition & Task Graphs
Self-Improving Eval-Gated Loops
Prompt Engineering
Retrieval & Data
Vector DBs (Pinecone / Chroma)
Embeddings & Reranking
Hybrid Search
Chunking Strategies
Synthetic Dataset Generation
ETL Pipelines
Fine-Tuning & Optimization
QLoRA / LoRA
PEFT
TRL
bitsandbytes
Distributed Training (DeepSpeed ZeRO)
Quantization
KV-Cache Compression
Adapter Merging
Training from Scratch
Hugging Face Hub
Inference & Deployment
Ollama
vLLM
llama.cpp
OpenAI-Compatible Endpoints
GGUF & Modelfiles
Chat Templating
Inference Benchmarking
Docker
FastAPI
Evaluation & Reliability
NLP
Model Evaluation (Evals)
Hallucination Mitigation
Guardrails
Observability & SLOs
Weights & Biases
Projects
Things I've built
sapiens
A Claude Code skill that explains AI/ML and LLM-engineering code clearly — to the point, in plain language, grounded in your codebase rather than generic snippets. The evolution of "caveman": where caveman compresses to save tokens, sapiens clarifies so the code is actually understood.
Claude Code
Agent Skills
LLM
omni
A durable, observable, self-improving agentic operating system you talk to — with zero dependencies. Omni turns natural-language requests into verified work: it decomposes goals into tasks, runs specialized workflows, keeps durable memory, governs risky actions, and improves itself behind an evaluation suite. Built entirely on the Node standard library.
Node.js
Agents
SQLite
DerpGPT
A conversational 50M-parameter AI model that generates responses from synthetic persona-chat datasets. Trained on multi-turn dialogues but often struggles to stay coherent — making up bizarre responses, misunderstanding context, or just plain derping out. It's here to entertain.
Python
PyTorch
Build your own AI model from scratch
An AI model built on the transformer architecture, made in Python with PyTorch. The code is also explained on my YouTube channel, where I made a series guiding viewers through the whole project.
Python
PyTorch
Gallery
Life in pictures
Contact
Let's build something
Feel free to reach out for collaborations or just a friendly hello.