LangGraph-based workflow engine powering 10+ legal automation scenarios. Supports contract drafting, comparison, review, compliance, and document inspection via tool calling and subgraph composition.
Projects
A selection of projects spanning LLM serving, legal AI, evaluation systems, and open-source contributions.
WebView-based AI workflow add-on for Hancom HWP document editor, enabling direct AI functionality within documents.
An ongoing partner product combining BoostDraft's Word-native editing interactions with allibee's near-human contract review depth — currently in active negotiation and co-development.
Demo apps, playbook QA tools, and workflow demo UIs for validating and showcasing legal AI capabilities.
4 PRs merged to predibase/lorax. Improved OpenAI-compatible serving interface, deterministic generation, streaming usage, and structured output support.
Multi-axis evaluation framework combining traditional NLP metrics, LLM-as-a-Judge, and legal expert rubrics. 20+ custom metrics, 10+ model benchmarks.
Government AI datacenter GPU infrastructure program. Led end-to-end from proposal writing and research execution to quarterly reporting and outstanding performance recognition.
End-to-end long-context work on legal documents of 50K–100K+ tokens: successful open-model SFT convergence at 100K+ context on a single GPU server node, a matching long-document evaluation pipeline, and production serving on top of the trained model.
End-to-end AI system for contract analysis covering extraction, checklist review, risk assessment, and revision suggestions. Patent registered.
End-to-end training pipeline for legal domain LLMs. Achieved SOTA on contract analysis benchmarks using SFT, DPO, and task-specific LoRA adapters.
NAVER Cloud Document Intelligence internship project. Built summary-based title generation for CLOVA Note and improved weak label quality via positive filtering SFT.
Graduation project combining INT8 quantization and LoRA adapters to fine-tune OPT-13B on a single consumer GPU (RTX 3090 24GB).
1st place in the NAVER CLOVA AI RUSH 2022 Long Document Summarization track. Domain-specific branched summarization system for book, dialogue, document, and journal inputs.
Developed security monitoring, infrastructure operations, and task automation programs in air-gapped environments as an InfoSec specialist in the Republic of Korea Army.
No projects in this category.