π AI/ML Enthusiast | π€ Infrastructure & Devops | π― Passionate about building real world solutions
I'm an enthusiastic as an Artificial Intelligence and Machine Learning engineer, my focus is on the convergence of Agentic AI and High-Performance Computing (HPC). I specialize in architecting intelligent, automated environments for complex physical systems. Currently, I am pioneering Project Trident, a modular CI/CD framework for the gprMax (GSoC '26) ecosystem. My work transcends traditional automation by integrating AI-driven validation gates that ensure 100% geometric and numerical integrity before simulation deployment.By leveraging Kubernetes (K8s) orchestration and Persistent Volume (PV/PVC) aggregators, I have engineered a distributed data pipeline that unifies high-fidelity simulation outputs from parallelized clusters. My approach utilizes Machine Learning principles to optimize resource allocation and 'Shift-Left' error detection, achieving sub-millimeter coordinate precision ($\pm 0.001$m) in the spatial domain. Whether I am fine-tuning LLM-based agentic workflows or scaling Physics-aware Docker containers, my goal is to transform manual, fragmented research processes into autonomous, production-grade systems that define the future of open-source engineering.
-
π Iβm currently working on open source contributer
-
π± Iβm currently learning I'm currently exploring Web Development to complement my AI/ML skills β learning how to build responsive, intelligent, and user-friendly web apps. From mastering HTML, CSS, and JavaScript to integrating backend logic with Flask and FastAPI, I'm focused on creating seamless experiences powered by smart models ,Go , Docker , python and machine learning , Spanner .
-
π― Iβm looking to collaborate on Keploy
-
π€ Iβm looking for help with Agent Swarm - Multi-Agent Orchestration
-
π¨βπ» All of my projects are available at https://v0-about-me-section-five.vercel.app/
-
π I regularly write articles on https://linktr.ee/pratiktech28
-
π¬ Ask me about Real-world AI Solutions, Keploy Architecture, and Scalable Backend
-
π« How to reach me pratiktech28@gmail.com
-
β‘ Fun fact I spend more time debugging Keploy PRs than sleeping! π€
π Engineering a Robust Validation Engine for gprMax @ GSoC 2026
---π οΈ Technical Arsenal & Core Competencies
| Domain | Tools & Technologies | Specialization |
|---|---|---|
| π°οΈ GSoC & DevOps | Docker, Kubernetes (K8s), GitHub Actions, Linux (Ubuntu) |
Automated CI/CD Pipelines & Scalable Infrastructure |
| ποΈ Physics & Simulation | gprMax, HDF5 Data, Numerical Accuracy Validation |
Deterministic Modeling & High-Fidelity Simulations |
| π€ AI & Machine Learning | Python (NumPy/Pandas), Agentic LLMs, LangChain, ElevenLabs |
Autonomous Workflows & Physics-Aware AI Gates |
| π» Full-Stack Dev | Node.js, React.js, Express, MongoDB, HTML5/CSS3 |
Scalable Web Architectures & Visual Data Dashboards |
| π Core Languages | C++, Python, JavaScript (ES6+), Bash Scripting |
Algorithm Design & System-Level Scripting |
| πΎ Cloud & Storage | Persistent Volumes (PV/PVC), GitHub Artifacts, AWS |
Distributed Data Persistence & Cloud-Native Storage |
π― Key Core Milestones Domain,Achievement,Tech Stack
GSoC Architecture,Project Trident: Modular CI/CD for gprMax,"GitHub Actions, Docker, Linux" Distributed Systems,Data Aggregator: K8s PV/PVC multi-pod sync,"Kubernetes, YAML, Bash" AIML & Data,Spatial Fidelity: Automated Physics-Aware Testing,"Python (NumPy/Matplotlib), RegEx" AI Automation,Vocal-Compute: Voice-First LLM Integration,"ElevenLabs, Gemini API, Node.js"
π§ AI Agents~
github.com/pratiktech28/ai_agents
Smart agent that analyzes patient data and suggests treatments using ML models.
π§ Bike_drivind_3d~
github.com/pratiktech28/Bike_driving_3d
π Faulty calculator in -js
github.com/pratiktech28/Calculator in js
password-gen.py
github.com/pratiktech28/password-generator.py
πΈπ ππππ£ππ€π₯π£π¦ππ₯π¦π£π πΌπππππππ£ | πΎππ β βπ ππ₯π£πππ¦π₯π π£ | π»πππ‘ πππππππ£
π¬ πππ₯'π€ βπ πππππ₯ & ππ ππ§π π€π ππ 'βπππ π€'!
I'm always open to discussing AI Infrastructure, Golang, or any interesting open-source project. Feel free to reach out!
- π§ πΌππππ: pratiktech28@gmail.com
- πΌ ππππππππ: https://linkedin.com/in/prateeksharma2807
- π βπ£π πππ€π€ππ πππ βππ€π¦ππ: [Download My Resume PDF]Black White Professional Minimalist Resume.pdf
"πΈ π»πππ‘ πππππππ£ ππ¦ππππππ π₯ππ πππ©π₯ πππππ£ππ₯ππ π π π© βππ π¦π-βππ₯πππ π€πͺπ€π₯πππ€."
---If you're into AI agents, creative coding, or social impact tech β let's build something amazing together π‘
π Passionate about building AI agents, smart web apps, or solving real-world problems with code?
Let's team up and turn ideas into impact! Whether it's a hackathon, open-source project, or just a fun experiment β
I'm always up for meaningful collabs π‘
Drop a message, share your vision, and letβs build something amazing together! πβ¨
I'm on a mission to blend AI/ML intelligence with creative web experiences. Whether it's building smart agents, deploying dashboards, or crafting clean code β I love turning ideas into impact. Let's connect, collaborate, and code something awesome together! π€π‘
Building High-Performance Physics-based CI/CD & Cloud-Native Simulation Infrastructure
| Project | Description | Tech Stack |
|---|---|---|
| π± Project Hub | Centralized GSoC '26 automation logic. | Python GitHub Actions |
| π Containerized gprMax | Scalable K8s/Docker simulation pods. | Kubernetes Docker |
| π‘οΈ gprMax Official | Main research documentation & source. | C++ OpenMP |
Deep-dive into technical validation of high-frequency electromagnetic environments.
- βοΈ Integration Testing: Automated E2E flows for simulation accuracy.
- π‘ A-Scan Processing: Signal validation logic & waveform analysis.
- π Geometry Check: Automated structural integrity verification.
- π§ͺ Material Validation: Real-time physics-property verification.
- π Interactive Portfolio: Full showcase of cloud & web architecture.
- π All-in-One Linktree: Unified access to all social & tech handles.
"Your work is the only signature that never fades."
Ziddi Engineering since 4 AM. π‘οΈ
Architecting Automated CI/CD & Physics-Based Regression for gprMax (GSoC 2026)
π‘οΈ Project Executive Summary: The Trident Pipeline (GSoC '26) The Trident Pipeline is a high-performance Containerized CI/CD Infrastructure specifically engineered to automate and validate complex physics simulations for the gprMax ecosystem (GSoC '26). The core architecture focuses on Physics-Aware Regression Testing, utilizing A-Scan Signal Processing and Numerical Stability checks to ensure 100% simulation accuracy after every code commit. By leveraging Docker and Kubernetes (K8s), the system achieves Elastic Scalability (10x to 100x), empowering researchers to transcend local hardware constraints and deploy simulations on enterprise cloud platforms like Google Kubernetes Engine (GKE) and Amazon EKS. Built on a robust GitHub Actions framework, this infrastructure guarantees Bit-for-Bit Identical Execution, effectively eliminating environmental drift and the "it works on my machine" fallacy. Furthermore, the integration of Apache Airflow and RabbitMQ enables asynchronous task management and complex simulation queuing, providing a Platform Agnostic Architecture for Newcastle University and the global research community. This project establishes a new benchmark for High-Performance Computing (HPC) in scientific software engineering, ensuring Sustained Contributor Commitment and long-term architectural integrity.




