AI/ML Engineer
- Engineered specialized AI agents and internal tools to optimize the recruitment lifecycle.
- Built automated candidate interview assessments and AI-driven sales-efficiency workflows.
Building LLM agents, RAG, and AI automation.
AI/ML Engineer at Gloroots · Bengaluru, India · Open to full-time, contract, and research-oriented AI roles globally.
I build AI/ML systems that combine product taste, LLM evaluation thinking, and enough engineering discipline to survive outside a notebook.
I’m an AI/ML Engineer at Gloroots in Bengaluru building hiring agents, coding assessments, internal AI tools, LLM automation, and product-facing workflows that have to hold up in front of real users.
My work sits at the overlap of model behavior, RAG, fine-tuning, LangGraph-style orchestration, evaluation loops, system design, and the small implementation details that make an AI product feel reliable instead of demo-only.
Projects that show how I think.
A mix of shipped experiments, benchmarks, and AI systems — written for what each one proves, not just what it uses.
Ayurvedic AI
RAG-based Ayurvedic medicine recommender using Gemini - won 1st Prize at MP Young Scientist Congress 2024.
GenAI Security System
Real-time security system bridging hardware (Arduino pressure plate) with Gemini Vision to detect suspicious activity.
GraphRAG AI Doctor
Graph-based knowledge system integrating Neo4j and LangChain for context-aware medical Q&A with efficient retrieval.
Multi-Agent Market Research
Multi-agent LangGraph framework where specialized agents collaboratively perform market research and ideate AI use cases with resource links.
TeleInteruv
Audio-based AI interview platform for conducting and evaluating technical interviews remotely.
UNO Arena
Designed an arena to benchmark LLM performance in adversarial UNO - testing reasoning, bluffing, and strategy.
Instruction Finetuned LLaMA 3.3B
Fine-tuned Llama 3.2 3B using PEFT (LoRA, gradient checkpointing) and Unsloth on 'The Tome' dataset, enabling effective multi-turn conversations.
LLaMA3
PyTorch
Unsloth
LoRA
Where the work has landed.
The through-line is pretty consistent: build AI systems that are useful, opinionated, and grounded in what people actually need.
AI/ML Intern
- Built an AI-powered coding assessment tool and AI interviewer to streamline technical hiring.
AI Intern
- Built generative AI workflows for studio-quality apparel imagery for The Souled Store.
- Used context engineering plus SD/Flux models to preserve brand consistency and reduce production time.
What I reach for when the problem is real.
The stack changes, but the pattern doesn’t: prototype fast, test the model logic, and keep enough engineering discipline that the thing can keep living after the demo.
Core lanes
LLM Agents
RAG
Fine-tuning
Computer Vision
Automation
Evaluation Loops
Languages
Python
JavaScript
C/C++
AI / ML
Transformers
LangChain
LangGraph
PyTorch
TensorFlow
OpenCV
Unsloth
Data / Infra
Neo4j
PostgreSQL
MongoDB
ChromaDB
AWS
Shipping layer
FastAPI
Django
Flask
Node.js
TailwindCSS
Pandas
NumPy
Scikit-learn
If you need AI work with taste and follow-through, say hi.
Best fit for roles or projects around agents, retrieval systems, internal automation, evaluation-heavy AI workflows, and product-facing ML systems.
Best places to find me
Availability