Backend & ML Engineer · Python · FastAPI · LLMOps · GenAI Systems

Software Engineer with 1+ year of experience building backend systems and ML-integrated services using Python, FastAPI, and Docker. Hands-on with RAG pipelines, LLM inference, NLP model development, and developer tooling. I prefer practical, production-adopted solutions over demos that don't ship.

📍 Based in Bengaluru, India

Nov 2025 – Present · Bengaluru, India

• Designed and maintained scalable REST APIs using FastAPI and Flask, supporting NLP and LLM-backed microservices consumed by downstream engineering teams.
• Integrated LLM inference endpoints and NLP preprocessing pipelines into backend services, reducing manual data processing overhead.
• Managed containerized service deployments using Docker across development and test environments.
• Modeled and queried structured and unstructured data using MySQL and MongoDB.
• Self-initiated and shipped an internal multi-service developer tools platform (FastAPI + Flask, 30+ REST endpoints) adopted organically by the team — spanning Jenkins CI tracking, async test report conversion, Apache log analytics, and Markdown documentation hosting.

Jun 2025 – Oct 2025 · Bengaluru, India

• Built REST APIs using FastAPI and Flask for ML inference and data processing features.
• Developed Python-based ETL and data processing scripts for NLP experimentation and model evaluation.
• Reduced model inference latency through batching strategies and INT8 quantization.
• Deployed and validated services on Docker containers and AWS EC2 instances.

Languages & Core

• Python 3.x, TypeScript, Bash
• REST APIs, Microservices, OOP, System Design

Backend Frameworks

• FastAPI, Flask, Django (basics)
• ASGI / WSGI, Jinja2, Gunicorn, flask-limiter
• Webex Bot Framework

ML & NLP

• PyTorch, TensorFlow, Hugging Face
• Fine-tuning, Embeddings, Seq2Seq & Attention
• CNN / GNN, Multi-Modal Fusion, Anomaly Detection

LLMOps & GenAI

• RAG, LangChain, LlamaIndex, Ollama
• Prompt Engineering, MCP, Agentic AI
• VS Code LM API

Databases

• MySQL, MongoDB, PostgreSQL (basics)
• Vector DBs: FAISS, Chroma

DevOps & Infrastructure

• Docker, Linux, Bash scripting, Nginx
• Git, Jenkins, CI/CD
• AWS EC2, Azure

Self-initiated, team-adopted multi-service productivity platform — covering CI monitoring, test reporting, log analytics, and documentation.

• Unified FastAPI + Flask platform with 4 ASGI-mounted sub-apps and 2 Flask frontends serving 30+ REST endpoints.
• Real-time Jenkins CI Progress Tracker with multi-stage regex log parsing and a Webex Teams bot (Adaptive Cards).
• Async HTML-to-Excel test report converter with background task queuing and styled multi-sheet workbook generation.
• Data-driven docs system rendering Markdown with TOC, Mermaid diagrams, and syntax highlighting.
• Apache log analytics dashboard with Chart.js visualizations and tail-based analysis.
• Unified Bash CLI (start/stop/backup/restore) with TUI, hash-based dependency tracking, and cron-based daily backups.

Tech: Python, FastAPI, Flask, ASGI, Webex Bot, Jenkins API, pandas, openpyxl, Bash

VS Code extension adding a persistent, workspace-aware memory layer to GitHub Copilot.

• Two-layer memory system (session logs + long-term store) for automatic context retention across sessions.
• Workspace-aware agentic coding via the VS Code Language Model API.
• Telegram Bridge for remote agent control with full tool access.
• Interactive Memory Browser webview, model switching, and JSON-based memory import/export.
• Docs · Marketplace · GitHub

Tech: TypeScript, VS Code Extension API, GitHub Copilot API, LLMOps, Agentic AI, MCP

GenAI-powered quiz generation platform using LLMs and retrieval-augmented generation.

• RAG pipeline combining embedding-based semantic retrieval with LLM-based question generation.
• Low-latency inference APIs using FastAPI for real-time quiz generation.
• Vector-based semantic search via FAISS / Chroma.
• Deployed via Docker + Nginx + Gunicorn.
• GitHub

Tech: Python, FastAPI, RAG, Embeddings, FAISS, LLMOps, Docker, Nginx

Full-stack Flask file sharing platform with user accounts, file bundles, expiration controls, admin dashboard, and PWA support.

• Drag-and-drop uploads, multi-file bundles, shareable quick-links, download tracking, and file expiration.
• Secure authentication with password hashing, email verification, password recovery, and CSRF-protected sessions.
• Admin dashboard with per-user storage quotas, server-wide limits, and SMTP configuration.
• Installable Progressive Web App with service worker.
• Demo · GitHub

Tech: Python, Flask, Flask-Session, Flask-Mail, Authentication, PWA, Pillow

Single-file Flask web app for sharing self-contained HTML pages and slideshow decks.

• HTML page sharing, source remixing, and multi-slide deck hosting with short-link routing.
• Full auth with email verification, password reset, and email-confirmed account deletion.
• Admin dashboard with tri-state freeze controls, pinned hero content, hit counters, and per-user activity logs.
• Rate limiting via flask-limiter, HTML sanitization, and async email delivery.
• Demo · GitHub

Tech: Python, Flask, REST API, Authentication, SMTP, Rate Limiting, Jinja2

Multi-modal DL system for automated defect detection in 3D mesh files.

• Fusion of pretrained ResNet CNN (76%) + 13-layer GNN (78%) → 85% fusion accuracy.
• Trained on ModelNet40 images and a custom proprietary OFF-file dataset.
• Graph construction from 3D mesh files using node features and adjacency matrices.
• GitHub

Tech: TensorFlow, CNN, GNN, Multi-Modal Fusion, ResNet, Trimesh

NLP translation system for code-mixed Tamil-English text using sequence-to-sequence deep learning.

• Seq2Seq models with attention mechanisms for code-mixed translation in PyTorch.
• Embedding-based sentence representations for cross-lingual transfer on low-resource pairs.
• Inference APIs exposed via Flask.
• GitHub

Tech: NLP, Seq2Seq, Attention, PyTorch, Hugging Face, Flask

Semantic log analysis using LLMs and embedding-based similarity search.

• FAISS vector similarity search for log anomaly detection and pattern clustering.
• LLM fine-tuning and prompt optimization for log summarization and root-cause inference.
• Real-time inference REST endpoints streaming insights into developer dashboards.

Tech: LLMOps, Embeddings, FAISS, Ollama, Python, Anomaly Detection

B.Tech, Artificial Intelligence and Machine Learning Kongu Engineering College, Perundurai · 2021 – 2025 · CGPA: 8.15 / 10

Relevant coursework: Natural Language Processing, Deep Learning, Machine Learning, Data Structures & Algorithms, Operating Systems.

Secretary — AI Coding Club · AI Department, Kongu Engineering College · 2022 – 2025

• Led 10+ hands-on workshops on NLP, ML systems, and Python-based AI development for a 100+ member student community.
• Mentored students through end-to-end ML and NLP project development, from ideation to deployment.

• Backend system design (FastAPI, Flask, async Python)
• RAG, LLM orchestration, and vector search
• Applied ML for detection and NLP problems
• Developer tooling, DX, and automation
• Docker, Linux, CI/CD

• 🌐 LinkedIn: linkedin.com/in/gdhanush270
• 💻 GitHub: github.com/gdhanush27
• ✉️ Email: gdhanush270@gmail.com

⭐ If you find any project useful, feel free to star it.