[SYS_INIT]ENGINEERING DOSSIER // ARPAN PRAMANIK
CLEARANCE: PUBLICINDEX: AP-2026
Arpan Pramanik

Arpan Pramanik

B.Tech CSE (AI & ML)

West Bengal, India
INSTITUTIONThe Neotia University
CORE FOCUSAI Agents & Full-Stack
DEPLOYMENTS10+ Production Systems
RESEARCH4 Conference Papers

ENGINEERING WITH RIGOR: AI AGENTS, RAG & PRODUCTION PLATFORMS.

I am an AI/ML Engineer and Full-Stack Developer specializing in building autonomous AI agents, intelligent retrieval systems (RAG), and production-grade web platforms. My focus centers on engineering scalable deep learning architectures and high-performance software that bridge scientific research with real-world utility.

From authoring PaperLens AI (an autonomous academic research co-pilot with hybrid FAISS+BM25 retrieval) to deploying distributed enterprise platforms like CampusSphere and explainable deep learning pipelines, I emphasize clean architecture, deterministic latencies, and visual polish.

Autonomous AI Agents

ReAct agent loops, streaming citations, tool execution, and prompt compression.

Grounded RAG Pipelines

In-memory BM25 + FAISS hybrid search paired with remote Supabase pgvector.

Production Full-Stack

Next.js 15, React 19, FastAPI, PostgreSQL, and distributed OpenTelemetry tracing.

[TECH_STACK]AUTHENTIC TOOL ECOSYSTEM

TECHNICAL ECOSYSTEM & CAPABILITIES.

Python
Python
Core AI Engine
PyTorch
PyTorch
Neural Architectures
🤗
Hugging Face
Transformers & LLMs
RAG
FAISS
Vector Similarity Indexing
AGENT
LangChain
ReAct Agent Loops
OpenCV
OpenCV
Computer Vision
[CHRONOLOGY]EXPERIENCE & INTERNSHIPS

PROFESSIONAL ROLES.

AI Research Intern

National Institute of Technology (NIT) Durgapur•Durgapur, West Bengal
June 2026 (4 Weeks)

Conducted research on lightweight deep learning architectures by optimizing EfficientNet variants for image classification tasks with a focus on model efficiency and inference speed.

Key Deliverables & Research:
  • Evaluated architectural variants across the EfficientNet family
  • Benchmarked performance on open-source image datasets
  • Analyzed trade-offs between accuracy, parameter efficiency, and FLOPs
  • Optimized deep learning models for efficient execution
PythonPyTorchTensorFlowEfficientNetDeep LearningComputer VisionModel Optimization
[VERIFICATION]CREDENTIALS & CERTIFICATIONS

ACCREDITATION.

🔒

Cyber Security Training Certificate

DataSpace Academy & The Neotia University • July 2024

☁️

Foundations and Practical Applications of AWS

DataSpace Academy & The Neotia University • November 2024

🎓

Revolutionizing Signal Processing: The Impact of AI and ML

DIT-IEEE-SPS • March 2025

🛡️

Cyber Security Advanced Training

Webel PVT. LTD & The Neotia University • June 2025

🐍

Flask Development Certification

Xetalabs • July-September 2025

🏆

Smart India Hackathon (SIH) 2025 — Internal Hackathon

The Neotia University & Ministry of Education, Govt. of India • September 2025

📰

Certificate of Merit — Anuranan

The Neotia University (Anuranan Journal) • February 2026