Dev Adhikari

AI Systems

End‑to‑End Design

Backend APIs

Production‑Ready

Agentic AI

Autonomous Workflows

Engineering Philosophy

I focus on building intelligent systems, not isolated models. My work blends backend engineering discipline with modern AI techniques, ensuring that machine learning models are deployable, scalable, observable, and useful in real-world applications. I prioritize system design, data flow, and long-term maintainability.

Advanced Technical Expertise

Backend & APIs

Django, FastAPI, REST, JWT/Auth, async APIs, service separation, AI‑first backend design.

Machine Learning

Classical ML, feature engineering, TF‑IDF, similarity systems, evaluation metrics.

Deep Learning

CNNs, transfer learning, VGG16 fine‑tuning, medical imaging pipelines.

LLMs & Agentic AI

LangChain, tool‑using agents, memory, planning loops, Hugging Face models.

RAG Systems

Embeddings, vector similarity search, document chunking, semantic retrieval.

AI Deployment

Model serving, FastAPI inference layers, integration with production backends.

Flagship Projects

Social Media Platform

Full‑stack Django application with authentication, feeds, scalable data models.

Chat with YouTube Videos

LLM‑powered conversational AI using transcripts, embeddings, and semantic search.

Brain Tumor Detection

Healthcare AI system using fine‑tuned VGG16 CNN on MRI datasets.

Movie Recommendation Engine

Content‑based recommender built using TF‑IDF vectorization and cosine similarity.

System Design Strengths

AI‑Backend Integration

Designing APIs that serve ML models reliably at scale.

Data Flow Thinking

Understanding how data moves from user input to model inference to output.

Future‑Proof Architecture

Building systems that can evolve as models and tools improve.