Skip to content

HexaForge-1/development_portfolio

Folders and files

NameName
Last commit message
Last commit date

Latest commit

ย 

History

12 Commits
ย 
ย 

Repository files navigation

Hi there, I'm Deepak Mishra ๐Ÿ‘‹

๐Ÿš€ C++ Software Developer | Systems Programming & AI Integration Enthusiast ๐Ÿ”ง Expert in High-Performance Computing, Memory Management & Multi-threading ๐Ÿค– Building scalable applications leveraging AI frameworks (TensorFlow C++ / PyTorch C++) ๐ŸŽฎ Experienced in Game Development with Unreal Engine & Real-time Simulation ๐ŸŽฎ Key domain areas - Firmware, Embedded, Middleware, Image processing, Cyber Security, Networking, Telecom etc. ๐ŸŒ Passionate about optimizing algorithms and writing clean, maintainable code ๐Ÿ“š Continuous learner | Open-source contributor | Problem solver


๐Ÿ”ง Skills & Technologies

  • Programming Languages:
    • C (ANSI C, Embedded C)
    • C++ (C++11, C++14, C++17, C++20) โ€“ Modern features like smart pointers, concepts, ranges, modules, coroutines, constexpr enhancements, and more
  • Databases:
    • Oracle, MySQL, PostgreSQL, SQLite
    • SQL query optimization, stored procedures, triggers
  • AI & Machine Learning Frameworks:
    • TensorFlow C++ API, PyTorch C++
  • Game Development:
    • Unreal Engine, OpenGL
  • Tools & Technologies:
    • Git, Docker, Visual Studio, CMake, Valgrind, GDB
  • Concepts & Techniques:
    • Multi-threading & concurrency (std::thread, async, mutex)
    • Memory management (smart pointers, RAII)
    • Design patterns, SOLID principles
    • Performance optimization and profiling

๐Ÿ“‚ Projects

  • ChronoMemDB A high-performance in-memory database engine written in C++17/20, featuring:

    1. Fast key-value storage and retrieval
    2. SQL-like query parser supporting basic commands (INSERT, SELECT)
    3. CMake build system for easy compilation
    4. Continuous Integration: Build & Test automatically via GitHub Actions on every pull request
      Repository: ChronoMemDB
  • FinWorkOptimizer : Enterprise Finance & Workflow Platform A Well defined integration of Finance, Workday, CRM & Oracle Fusion workflows with payroll, expenses, invoicing, budgeting, forecasting, and secure GL exports for enterprise growth, featuring:

    1. REST-based system integrating Finance, Workday, CRM & Oracle Fusion-style operations.
    2. Core Modules: Payroll, Expenses, Invoicing, Timesheets, Employee Management.
    3. CRM Features: Customer management (add/update, list).
    4. Workflow Approvals: Timesheet approval endpoint with secure API key authentication.
    5. Budgeting Tools: Set/update budgets, variance analysis reports.
    6. Forecasting: Cashflow forecasting with moving average.
    7. Oracle Fusion GL Export: Generate CSV-ready files for import.
    8. Security: API key authentication, configurable via environment variables.
    9. Enterprise Ready: Scalable architecture, modular design, CI/CD with build & test enforcement.
      Repository: FinWorkOptimizer
  • HighPerf-AI-ML-Inference A high-performance C++ AI/ML inference engine built using ONNX Runtime (CPU by default, GPU opt-in) with optional LibTorch support. This CLI + REST API tool is cross-platform, powered by CMake, and deployable via Docker and GitHub Codespaces. It auto-fetches models/assets, includes a CUDA runtime check, and enforces CI-based merge gating. Future plans include multi-model and GPU training support.

    1. Unified CPU/GPU inference โ€” pick runtime flexibly
    2. CLI & REST interfaces for integration versatility
    3. Seamless asset/model fetching and environment validation
    4. Cross-platform builds with CMake and Docker
    5. Maintains high quality through CI gating on PRs Repository: highperf-ai-ml-inference

๐Ÿ“ซ Contact Me


โšก GitHub Stats

https://github-readme-stats.vercel.app/api?username=HexaForge-1&show_icons=true&theme=radical


"Code is like humor. When you have to explain it, itโ€™s bad." โ€“ Cory House

About

GitHub profile README presenting my work, technologies, and developer journey.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors