Primary node
GLA University
Bachelor of Technology in Computer Science Engineering
August 2019 to August 2023
Chamber · India
Systems online
UT Space
00 / Chamber
Full-stack and AI-focused software engineer building modern web applications, AI-powered systems, and developer-focused products.
React and Next.js, backend engineering, LLMs, RAG, MCP, and high-performance interfaces.
Core stack orrery · React · Next · FastAPI · RAG · MCP
Scroll to inspect identity, formation, systems, and instruments.

Utkarsh Ojha · India
01
Understand the problem deeply before choosing the technology.
02
Prefer clear, scalable architectures over unnecessary complexity.
03
Design for real-world operation, not just demonstrations.
04
Treat performance, maintainability, UX, and developer experience as a single problem.
05
Use AI to accelerate engineering while remaining fluent in the systems underneath.
06
Stay in motion from interface design to distributed systems, databases, and intelligence.
Full-stack engineer with deep work in React and modern web systems.
Moves across frontend, backend, data, cloud, and AI engineering.
Builds products, not isolated features.
Focused on LLM applications, RAG, MCP, agents, embeddings, and developer tools.
Explores mobile platforms and lower-level systems to understand what sits beneath the stack.
Chooses technically demanding work that reveals real engineering depth.
Primary node
Bachelor of Technology in Computer Science Engineering
August 2019 to August 2023
Satellite milestone
Full-stack engineering internship
Applied full-stack engineering in an academic research environment — working across the web stack, from interface to implementation.
Layer · Frontend
Apr 2023 — Present
Current post
Audax Labs
Engineering across web, mobile, APIs, and cloud — contributing to enterprise data systems and remote IoT firmware delivery.
Recognition
Audax Star Award
Recognized for contribution at Audax Labs.
An enterprise platform for data discovery, search, governance, and compliance. It reaches cloud and on-premise sources and provides search, classification, analytics, compliance, and GenAI capabilities.
Contributed to improvements associated with approximately 30% better data accessibility and compliance.
Remote firmware management for IoT devices — secure delivery, monitoring, and the operational path for keeping devices current without unnecessary downtime.
Contributed to a solution associated with approximately 25% reduction in device downtime.
04 instruments

RAG, agents, and MCP over a living codebase
An AI-powered developer assistant that connects to a GitHub repository, understands the codebase, and answers questions about architecture, implementation, defects, and specific modules.
Open case
Presence, sync, and shared state across instances
A real-time platform where multiple people can work on documents and projects at once — with live updates, presence, comments, and a shared workspace.
Open case
Multi-tenant retrieval with source-grounded answers
A multi-tenant AI knowledge platform that lets people upload and connect documents, repositories, and other sources, then ask questions in natural language and receive answers grounded in those sources.
Open case
A small PaaS: repository in, public URL out
A lightweight deployment platform where a developer connects a GitHub repository and the system builds, deploys, and exposes the application on a generated public URL.
Open caseSelf-assessed fluency on a 1–10 scale. The point is honesty: where the work is strong, and where it is still being sharpened.
LLMs
0/10
Prompting, API integration, structured outputs, streaming, and production-oriented LLM workflows.
RAG
0/10
Document ingestion, chunking, retrieval pipelines, context construction, and grounded generation.
MCP
0/10
MCP servers, tools, resources, and LLM-to-tool integrations.
AI Agents
0/10
Tool calling, agent workflows, context management, and task-oriented automation.
Embeddings
0/10
Semantic representations, embedding generation, and similarity-based retrieval.
Vector Search
0/10
Vector databases, similarity search, metadata filtering, and retrieval optimization.
Voice AI
0/10
Speech-to-text, Faster Whisper, intent classification, and text-to-speech pipelines.
Certification
PL-300
Marks in the record
The chamber stays open. Reach through GitHub or LinkedIn — the work is in India, the channel is wherever you are.