Available for opportunities

Architecting
resilient systems for a scalable future.

Hey, I'm Lagnajit Moharana. Full-stack developer exploring the frontier of AI & Machine Learning, with a foundation in Cloud Infrastructure and DevOps.
App Development Analyst | @Accenture.

Lagnajit Moharana
Core Disciplines

Full Stack

Crafting clean, responsive front-ends and robust back-end architectures with real-time capabilities.

DevOps

Deploying scalable applications with containerization, CI/CD pipelines, and infrastructure automation.

Machine Learning

Studying transformer architectures and building production systems with RAG pipelines and agentic workflows.

About

Who I am

I build end-to-end systems — from clean interfaces to scalable infrastructure.

I found my passion for coding back in college, diving headfirst into full-stack development. Since then I've built a range of end-to-end applications — from real-time collaboration tools to payment systems.

I started with front-end, crafting clean and responsive interfaces. Over time my focus grew into Cloud Infrastructure, DevOps, and AI-driven systems — designing and deploying scalable applications with containerization, CI/CD pipelines, and infrastructure automation.

These days, I'm deep into ML and AI — studying the internals of large language models, from transformer architecture and attention mechanisms to advanced concepts. I've trained and deployed my own models, and built production systems with RAG pipelines, agentic workflows, and tool use.

Away from the keyboard, you'll find me in the world of 2D animation, portrait drawing, and design — bringing characters to life and sketching intricate portraits.

Experience

Where I've worked

  1. Packaged App Dev Analyst

    June 2026Present

    Accenture, India

    Building and deploying resilient infrastructure automation platforms and ServiceNow workflows across client environments to optimize operations and maintain high availability.

  2. Packaged App Dev Associate

    August 2024May 2026

    Accenture, India

    Developed and maintained full-stack internal applications and optimized cloud operations utilizing Python, Django and Azure OpenAI.

Education

Where I studied

2021 — 2024

Ravenshaw University

B.Sc. in Information Technology & Management (87.13%). Built several full-stack projects spanning React.js front-ends, Node.js / MongoDB back-ends, and cloud services like Firebase — laying a solid foundation for my career.

2019 — 2021

Ravenshaw Higher Secondary School

12th, CHSE Board (94%).

2008 — 2019

Saraswati Vidya Mandir

10th, BSE Board (91.67%).

Credentials

Certifications

Technical Arsenal

Tech stack for high performance

Continuous exploration of the best tools to solve complex architectural challenges.

Cloud & DevOps

AWSAWS
AzureAzure
GCPGCP
DockerDocker
KubernetesKubernetes
Ansible
Ansible
TerraformTerraform
GitGit
GitHub
GitHub

AI & Machine Learning

PyTorchPyTorch
Hugging FaceHugging Face
T
Transformers
F
Fine-Tuning
L
LoRA
LangChainLangChain
Agentic AIAgentic AI
RAGRAG

Frontend

ReactJSReactJS
NextJS
NextJS
TypeScriptTypeScript
JavaScriptJavaScript
Tailwind CSSTailwind CSS

Backend

ExpressJSExpressJS
PythonPython
DjangoDjango
FastAPIFastAPI
RedisRedis
Prisma
Prisma
Selected Work

Projects

ReactJSDjangoRedisCeleryGCPDockerAgentic AIRAG

AutoSage: Infrastructure Automation Platform

AutoSage is an AI-driven infrastructure automation platform that streamlines and optimizes cloud operations through workflows, RAG and agentic AI, enabling users to manage and scale their infrastructure efficiently.

ReactJSExpressJSPostgreSQLTailwind CSSPrismaDockerMicroservicesKubernetes

SwiftPay: Payment Application

SwiftPay is a demo payment application that allows users to send and receive money instantly, connect with friends, and manage their finances with ease. It is built with a modern tech stack and offers a seamless user experience.

Learning in Public

ML Journey & Projects

Building and aligning language models from the ground up — each version pushing further than the last.

v1PyTorchGradioHuggingFaceTransformers & Attention

MiniGPT-Shakespeare

Where it started — a GPT-style transformer trained entirely from scratch in PyTorch on the works of Shakespeare, then deployed on HuggingFace Spaces with a Gradio interface for interactive text generation.

v2PyTorchGradioTransformers & AttentionRMSNormSwiGLURoPEGQABPE Tokenization

MiniGPT-v2

A GPT-style transformer built from scratch in PyTorch and trained on the TinyStories dataset. It layers in modern techniques — RMSNorm, SwiGLU, RoPE, GQA, LR scheduling and a hand-rolled BPE tokenizer — to improve quality and efficiency in text generation.

v3PyTorchTransformersStreamlitSmolLM-135MLoRASFTDPOGRPO

MiniGPT-v3

A fine-tuned SmolLM-135M taken through a complete alignment pipeline — Continued Pre-Training (CPT) on WikiText, Supervised Fine-Tuning (SFT) on Alpaca, preference alignment via RLHF/DPO on Orca pairs, and RLVR with GRPO for verifiable arithmetic reasoning. Each stage is a stackable LoRA adapter, served on Streamlit.

Contact

Let's build something extraordinary.

Have a project in mind, a role to fill, or just want to say hi? My inbox is always open — I'll try to get back to you within 24 hours.

2004lagnajitmoharana@gmail.com
Odisha, India — 753004