Software Engineer | AI Systems & Backend Architect
I specialize in building high-performance backend architectures and researching autonomous AI systems. My work bridges the gap between scalable distributed systems and cutting-edge Multimodal AI (VLMs/SSMs).
- Backend & Systems Engineering: Architecting complex document workflow orchestration, automating multi-cloud data exports, and optimizing high-traffic database schemas.
- AI & Research: Engineering Vision-Language-Action (VLA) pipelines and Multi-Agent Orchestration (MARS) with a focus on real-time risk modeling and edge-deployment.
- Security & Scale: Implementing zero-trust policy gates, OAuth 2.0/JWT integrations, and schema-driven development (Pydantic/Zod).
| Focus Area | Technologies |
|---|---|
| Backend | Python (FastAPI/Flask), MongoDB, Redis, Celery, Kafka, REST APIs |
| AI/ML Research | PyTorch, Transformers, Phi-4, CLIP, State Space Models (SSMs) |
| Agentic AI | LangGraph, LangChain, MCP Protocol, Multi-Agent Systems |
| Infrastructure | CI/CD, ONNX Runtime, TensorRT, Linux, GitHub Actions |
- Multimodal Perception: Exploring the intersection of Vision-Language Models and real-world robotics/action.
- Edge Intelligence: Optimizing heavy LLM/VLM workloads for low-latency inference on edge hardware via TensorRT.
- Autonomous Agents: Designing self-correcting agentic workflows for research and automation.
- 📧 paribeshpanta44@gmail.com
- 📍 Kathmandu, Nepal | 🇳🇵