I build AI systems that hold up in production — the unglamorous part where a demo becomes something a city actually runs on. I'm the primary engineer on four live government platforms at Gractor, a smart-city AI company: RAG over 2.7M+ sensor records at 98.8% accuracy, tool-calling agents, and the evaluation harnesses that keep them honest.
Earlier, during my MSc in AI at Korea University's Pattern Recognition & Machine Learning Lab (advised by Prof. Seong-Whan Lee, IEEE Fellow), I published two first-author papers on semi-supervised semantic segmentation, ranked #2 and #3 globally on the Cityscapes and Pascal VOC leaderboards, work I now continue independently with PixCon.
My focus: reliable retrieval, calibrated uncertainty, and closing the gap between a number that works in a paper and a system real users depend on.
| Year | Paper | Venue | Rank | Links |
|---|---|---|---|---|
| 2026 | PixCon — Clean-Positive Contrastive Learning for Foundation-Model SSSS | Preprint (under review) | #2 | arXiv · Code · Project |
| 2026 | FARCLUSS — Fuzzy Adaptive Rebalancing & Contrastive Uncertainty Learning for SSSS | Neural Networks (under review) | #2 | arXiv · Code · Project |
| 2025 | CW-BASS — Confidence-Weighted Boundary-Aware Learning for SSSS | IJCNN (IEEE) | #3 | IEEE · arXiv · Code · Project |
With ResNet-101 backbones, CW-BASS and FARCLUSS were among the state of the art in semi-supervised segmentation in 2025, ranking #3 and #2 globally on Pascal VOC / Cityscapes (77.15% and 78.8% / 78.2% mIoU). PixCon (independent, 2026) carries the line onto foundation-model features (DINOv2-scale), reaching #2 in semi-supervised segmentation, matching a strong UniMatch V2 baseline at lower cost. All three attack the same problem, dense prediction from very few labels: confidence-weighted boundaries (CW-BASS), fuzzy pseudo-labeling with contrastive rebalancing (FARCLUSS), and a clean-positive pixel memory bank (PixCon).
| Project | What it does | Stack |
|---|---|---|
| hwpkit | Read, fill & edit Korean HWP (Hancom Office) docs in Python — text extraction for LLM/RAG, programmatic form-filling, and corruption-free binary rewrite | Python · OLE/CFB |
| Claude Usage Widget & Token Tracker | Live system-tray widget for Claude Code plan limits (5h/7d) + local token & cost analytics per project, model, and tool | Python · GTK · CLI |
Claude Code usage — I build with agentic coding daily.
Local Claude Code telemetry, snapshot updated Jul 2026.
AI / LLM Systems
-6AA6FF?style=flat-square){kind=icon}
ML / Computer Vision
Backend & Data
Frontend
Infra / DevOps / IoT
Languages
AI/ML Engineer · Gractor Co., Ltd. · Sept 2025 – present · Seoul
Primary engineer across four live government platforms. Built a RAG system over 2.7M+ IoT records at 98.8% eval accuracy, rebuilt a production agent from 94.9% → 100% (96/96) with ~30% less code and ~12x faster startup, shipped a multi-provider LLM router with circuit-breaker failover, and deployed YOLOv5 + OpenVINO edge inference on government smart poles.
Research Engineer (MSc) · Korea University, PRML Lab · Sept 2023 – Feb 2026 · Seoul
Advised by Prof. Seong-Whan Lee (IEEE Fellow). First-author segmentation papers (#2 & #3 globally); ~10K LOC of PyTorch multi-GPU training infrastructure; Korean patent filed (autonomous-driving perception).
AI Software Engineer · GliT (GLITEC), EdTech · Jan 2019 – Jan 2021 · Zimbabwe
Built two offline-first mobile learning products reaching 500+ students and 80,000+ learning sessions.
MSc in Artificial Intelligence · Korea University · 2023–2026 · GPA 3.78/4.0
Global Korea Scholarship (sole Zimbabwe awardee) · BK21 Research Fellowship · Advisor: Prof. Seong-Whan Lee
Awards — GINCON Global Award 2025 (Korean National Assembly)
Let's build something that ships.
Like this profile? Reuse the design with attribution to Ebenezer Tarubinga. See LICENSE.