I build real, production-grade systems — not toy projects.
My strength is at the intersection of AI/ML, backend engineering, and SaaS product development.
I can architect a platform from scratch, integrate multi-agent LLM workflows, build scalable APIs, handle cloud deployments, and design systems that actually work under load.
👉 Live: https://issuenix.com/
Stack: FastAPI, PostgreSQL, React.js, Docker, JWT Auth
What I built:
- Backend architecture + DB schema
- Authentication, authorization, RBAC
- Certificate generation & secure verification engine
- File processing + storage pipelines
- Full frontend + backend integration
- Deployment + CI/CD
- AI assistant Sonia (Gemini + RAG) for workflow automation
This is a complete SaaS platform — built end-to-end.
👉 GitHub: https://github.com/Pankaj4152/smart-photo-finder
Stack: Python, Sentence-Transformers, FAISS, Pydantic, Embedding Pipelines
What it does:
A lightweight, offline-first semantic image search engine that retrieves images based on meaning, not filenames.
What I built:
- Image embedding pipeline
- Vector search engine for fast similarity lookups
- Batch image processing + folder-level embedding workflows
- Modular architecture.
- Extensible design (can become a FastAPI service or desktop/mobile backend)
- Fully offline, privacy-first photo organization system
An end-to-end AI utility built on embeddings + vector search.
👉 GitHub: https://github.com/Pankaj4152/AI-Insight-Pro
Stack: Python, FastAPI, BigQuery, MySQL, Postgres, MongoDB, ChromaDB
Highlights:
- Multi-database ingestion engine
- ETL pipelines + metadata extraction
- Query execution service
- Cloud integrations (AWS/GCP/Azure)
- RAG pipeline for policy/reports
- Multi-agent LLM workflows (Gemini)
Backend-heavy, platform-focused, with optional LLM features.
👉 GitHub: https://github.com/Pankaj4152/mylibrary
Stack: PyTorch, HF Transformers, FastAPI
Highlights:
- High-performance text generation & classification
- Thread-safe loaders, GPU cleanup, caching
- Shared model pools for speed
- Reusable internal modules
Production-grade inference system — your strongest ML systems project.
| Category | Skills |
|---|---|
| Languages | Python, C/C++, SQL |
| Backend / Platform | FastAPI, Flask, PostgreSQL, MySQL, MongoDB, BigQuery, REST APIs, Auth, RBAC, Caching, Concurrency |
| AI / ML / GenAI | PyTorch, TensorFlow, HuggingFace, Transformers (T5, DistilBART), RAG (ChromaDB), Multi-Agent LLM Workflows (Gemini), PII/SDE Pipelines |
| Cloud & DevOps | Docker, GitHub Actions, AWS, GCP, Linux |
| Frontend (for SaaS) | React.js, TailwindCSS |
| Other | Agile/SCRUM, System Design Basics, CI/CD, Git |
- AI-augmented products
- Backend & platform systems
- SaaS applications with real users
- Scalable APIs + ingestion pipelines
- Multi-agent LLM tools
- Developer tools & internal infra
I build fast, I ship real products, and I understand both AI and backend deeply.
If that’s the kind of engineer you need — let’s talk.