Turning ideas into production-grade software
- 🎓 M.S. Computer Science @ George Mason University - GPA 4.0 | Dean's List Fall '24 & Spring '25 | Graduated May 2026
- 🔬 Research Assistant under Dr. Duoduo Liao - deep learning for subsurface concrete defect detection using USW, IE, GPR & HCP signals
- 🔬 Research @ GMU — multi-agent text-to-SQL benchmark across OpenAI / Anthropic / Ollama models.
- 🤖 AI Engineer @ MITRE-Sponsored Capstone - built LLM-powered insider-threat simulation using NVIDIA LLaMA-3.1-Nemotron-70B
- 🏗️ Led a 7-member team at Xebia, shipping a full-stack app with 30% faster loads and 95% automated test coverage
- 🌍 CS UK Study Abroad alum - explored AI governance, cryptography, and global tech systems through an ethics-forward lens
LLM-based synthetic data generation for cybersecurity threat simulation
- Fine-tuned NVIDIA LLaMA-3.1-Nemotron-70B-Instruct to generate malicious, medical insider-threat, and neutral tweets with high contextual authenticity
- Drove Agile delivery with weekly client syncs, iterative user story refinement, and a non-technical-friendly Tkinter GUI
- Stack:
PythonLLaMA 3.1Hugging FacePrompt EngineeringTkinter
Deep learning framework for structural health monitoring of concrete infrastructure
- Training ResNet1D, CNN1D, and BiLSTM models on FHWA lab slab data to classify delamination, voids, cracks, and honeycombing
- Exploring multimodal data fusion — integrating USW, IE, GPR, and Half-Cell Potential (HCP) signals for robust NDE
- Stack:
PythonNumPyscikit-learnResNet1DBiLSTM
Production-grade platform with a fully automated CI/CD pipeline to Kubernetes on AWS EKS
- Engineered GitHub → Jenkins → Docker → AWS EKS pipeline with blue/green deployments
- Achieved 80% faster release cycles and 99.9% uptime across a 3-node cluster
- Stack:
AngularSpring BootMySQLDockerKubernetesJenkinsAWS
Real-time item management platform with secure auth and high test coverage
- Built real-time search, Firebase Google Auth, and scalable data storage with MongoDB
- Achieved 95% automated test coverage via Selenium + TestNG pipelines
- Stack:
React.jsMongoDBNode.jsFirebaseSelenium
🐱 My GitHub Data
📦 ? Used in GitHub's Storage
🏆 758 Contributions in the Year 2026
🚫 Not Opted to Hire
📜 28 Public Repositories
🔑 0 Private Repositories
I'm an Early 🐤
🌞 Morning 76 commits █████░░░░░░░░░░░░░░░░░░░░ 20.16 %
🌆 Daytime 160 commits ███████████░░░░░░░░░░░░░░ 42.44 %
🌃 Evening 95 commits ██████░░░░░░░░░░░░░░░░░░░ 25.20 %
🌙 Night 46 commits ███░░░░░░░░░░░░░░░░░░░░░░ 12.20 %
📅 I'm Most Productive on Thursday
Monday 31 commits ██░░░░░░░░░░░░░░░░░░░░░░░ 08.22 %
Tuesday 50 commits ███░░░░░░░░░░░░░░░░░░░░░░ 13.26 %
Wednesday 41 commits ███░░░░░░░░░░░░░░░░░░░░░░ 10.88 %
Thursday 72 commits █████░░░░░░░░░░░░░░░░░░░░ 19.10 %
Friday 48 commits ███░░░░░░░░░░░░░░░░░░░░░░ 12.73 %
Saturday 68 commits █████░░░░░░░░░░░░░░░░░░░░ 18.04 %
Sunday 67 commits ████░░░░░░░░░░░░░░░░░░░░░ 17.77 %
📊 This Week I Spent My Time On
🕑︎ Time Zone: America/New_York
💬 Programming Languages:
Markdown 2 hrs 8 mins ████████████░░░░░░░░░░░░░ 47.45 %
Kotlin 1 hr 14 mins ███████░░░░░░░░░░░░░░░░░░ 27.54 %
Python 37 mins ███░░░░░░░░░░░░░░░░░░░░░░ 13.94 %
Other 12 mins █░░░░░░░░░░░░░░░░░░░░░░░░ 04.78 %
Prolog 8 mins █░░░░░░░░░░░░░░░░░░░░░░░░ 02.98 %
🔥 Editors:
Claude Code 4 hrs 8 mins ███████████████████████░░ 91.97 %
VS Code 21 mins ██░░░░░░░░░░░░░░░░░░░░░░░ 08.03 %
🐱💻 Projects:
Lumen 2 hrs 29 mins ██████████████░░░░░░░░░░░ 55.47 %
Portfolio-Scraper 1 hr 56 mins ███████████░░░░░░░░░░░░░░ 42.96 %
alpha-peptide-labs 4 mins ░░░░░░░░░░░░░░░░░░░░░░░░░ 01.58 %
💻 Operating System:
Windows 4 hrs 30 mins █████████████████████████ 100.00 %
🤖 AI Coding This Week
⏱ AI Coding Time: 4 hrs 26 mins (98.63%)
✍️ 9,867 lines written by AI, 1 lines written by hand (99.99% AI-written)
🔤 4,243,772 Input Tokens, 484,482 Output Tokens
💵 $214.97 Estimated AI Cost This Week
🧠 9 AI Sessions, 51 AI Prompts
Opus 7,441 lines ███████████████████░░░░░░ 75.37 %
Fable 2,432 lines ██████░░░░░░░░░░░░░░░░░░░ 24.63 %
Claude-Code 0 lines ░░░░░░░░░░░░░░░░░░░░░░░░░ 00.00 %
🔎 AI Coding Insights:
🤖 AI-Driven — 99.99% of written lines came from AI
📚 Verbose Prompter — average 3,794 characters per prompt
🔁 Iterative Prompter — average 6 prompts per session
🚀 High AI Trust — 0.14% of changed lines were hand-edited
I Mostly Code in Python
Python 7 repos ████████░░░░░░░░░░░░░░░░░ 31.82 %
JavaScript 5 repos ██████░░░░░░░░░░░░░░░░░░░ 22.73 %
TypeScript 3 repos ███░░░░░░░░░░░░░░░░░░░░░░ 13.64 %
Jupyter Notebook 2 repos ██░░░░░░░░░░░░░░░░░░░░░░░ 09.09 %
Kotlin 1 repo █░░░░░░░░░░░░░░░░░░░░░░░░ 04.55 %
Timeline
Last Updated on 05/09/2026 03:51:51 UTC
- 🚀 Published release PDF Translator 1.0.0 in royalgillz/pdf-translator
- ❗ Opened issue #23540 in JonathanGin52/JonathanGin52
- ❗ Opened issue #23535 in JonathanGin52/JonathanGin52
- ❗ Opened issue #23533 in JonathanGin52/JonathanGin52
- ❗ Opened issue #23532 in JonathanGin52/JonathanGin52
- 🏆 Dean's List — Fall 2024 & Spring 2025 @ George Mason University
- ☁️ AWS Academy Graduate — Cloud Foundations Certified
- 🔮 Oracle AI Vector Search Certified Professional
- 🥇 PatriotHacks CTF @ GMU — Top 7 Finish
- 🥈 Hacklipse Hackathon by ACM TIET — Top 15 Finish
| Platform | Link |
|---|---|
| linkedin.com/in/sehaj-gill | |
| sehajgill808@gmail.com | |
| 🐙 GitHub | github.com/royalgillz |