Hi, I'm Yashvardhan Goel

|

Building intelligent systems that bridge the gap between research and real-world applications. Currently crafting AI solutions at Mira, Aroha Labs.

Get in Touch View Projects
knowledge graph
Scroll
About Me

Turning Complex Problems into Elegant Solutions

I'm an AI/ML Engineer with a unique blend of biological sciences and manufacturing engineering background from BITS Pilani. This interdisciplinary education gives me a distinctive perspective on solving complex problems.

My journey started in research at the intersection of biology, material science, and AI/ML, working on computational discovery of battery materials and mining biological literature using NLP and deep learning.

From there, I gained industry experience across diverse domains: healthcare (breast cancer classification at APS Labs), digital advertising (optimizing ad engagement at Rocketium), and FMCG at Hindustan Unilever Ltd. where I collaborated with Deloitte consultants on ML-driven process optimization. Currently at Mira, Aroha Labs, I build production-grade AI agents and event-driven microservices on AWS, powering intelligent systems with Agno, LangChain, vector databases, and MCP integrations.

Dual Degree M.Sc. Biological Sciences + B.E. Manufacturing
International Experience Singapore & Taiwan Research
Research Fellow SPARKLE & TEEP Scholar

Technical Arsenal

Languages & Frameworks

Python FastAPI LangChain Agno PyTorch

AI/ML

Agentic AI LLMs NLP Deep Learning Computer Vision

Infrastructure

Docker AWS GitHub Actions Nginx Redis

Databases

Milvus Pinecone FAISS PostgreSQL
Career Journey

Where I've Made an Impact

Software Developer - AI/ML

Mira, Aroha Labs
Singapore Nov 2025 - Present
  • Building production-grade AI agents using the Agno framework with custom memory architectures and Milvus-powered hybrid retrieval pipelines for improved reasoning and context-aware responses
  • Developing a learning pipeline that enables AI agents to continuously improve through an experiential feedback loop, leveraging Natural Language Inference (NLI) and Human-in-the-Loop (HITL) systems
  • Engineered integrations with enterprise tools (Google Workspace, M365, Hubspot, etc.) using MCP-compatible tool definitions, enabling AI agents to perform cross-platform actions
  • Architected event-driven microservices on AWS with SQS/SNS, ECS clusters, ALB, and VPC-isolated services
  • Set up CI/CD pipelines via GitHub Actions for automated deployment across all microservices on AWS
Agno Milvus NLI MCP AWS SQS/SNS

AI Engineer Intern

Mira, Aroha Labs
Singapore May 2024 - Oct 2025
  • Spearheaded the AI backend for Delphi Intelligence's chatbot using LangChain, Milvus, hybrid-retrieval, advanced memory, and intelligent routing for superfast, context-aware responses
  • Pioneered Mira Flows, a plug-and-play AI platform with 5,000+ users using microservice architecture
  • Built a Web3 security agent leveraging GoPlus APIs, utilizing Model Context Protocol (MCP) for real-time, automated security intelligence
LangChain Milvus MCP Web3

Machine Learning Intern

Rocketium, Mycelium Technologies
Bengaluru, India Jan 2024 - Apr 2024
  • Engineered ML model development by automating data extraction from Google Sheets with Google Apps Script
  • Developed fine-tuned U2-Net model for analyzing digital ads achieving 50% increase in engagement
  • Conducted comparative analysis of Azure, AWS Rekognition and Google Vision for image tagging
  • Boosted user engagement by 50% by optimizing ad strategies with fine-tuned ChatGPT and LlamaIndex
U2-Net LlamaIndex Computer Vision Ad-Tech

Machine Learning Intern

Hindustan Unilever Ltd. (HUL)
Assam, India May 2023 - Jul 2023
  • Collaborated with a 'Big Four' firm on ML-driven labor allocation, saving 1.5 hours daily
  • Employed GBM regressor for quality checks, reducing BCT by 20 minutes and boosting production by 50%
  • Spearheaded ML model for energy sustainability aimed at reducing carbon footprint
GBM Process Optimization Sustainability

Machine Learning Intern

APS Labs, Part of APS Lifetech
Pune, India May 2022 - Jul 2022
  • Leveraged ML and image processing to develop breast cancer classification system using mammograms
  • Evaluated multiple models, concluded SVM outperformed others with 95% accuracy
Healthcare AI Image Processing SVM
Highlighted Projects

A Few Things I've Built

Nov 2025 - Present Personal Project

TL;DAI (Too Long; Did AI)

Android app that aggregates AI and tech updates from arXiv, Reddit, GitHub Trending, and Hacker News. Features an AI summarization pipeline, bookmarking, and privacy-first local storage.

Android AI Summarization arXiv Hacker News
May 2025 - Jun 2025 Personal Project

Pokebase MCP Server

MCP server integrating Gemini 2.0 and PokeAPI for Claude Desktop with async HTTP client processing 1000+ Pokemon and intelligent battle strategy generation.

MCP Gemini 2.0 aiohttp NLP
Mar 2025 Personal Project

DermaGPT

RAG chatbot with LangChain, GPT-4, and FAISS/Pinecone for precise skincare product recommendations. Features dynamic chunking and real-time search integration.

LangChain GPT-4 FAISS Streamlit
May 2023 - May 2024 Research

Computational Discovery of Battery Materials

Adapted deep learning frameworks to predict chemical compound properties with 90% accuracy. Built BERT-based text extraction tool and employed hierarchical GNNs for novel material discovery.

BERT GNN Deep Learning Materials Science
Mar 2023 - Dec 2023 Research

Tool for Mining Biological Literature

Extracted research papers featuring rice-associated genes using NLP and PubMed's API. Designed a database-driven website to catalog genes and their isolation methods.

NLP BERT PubMed API Bioinformatics
Recognition

Achievements & Certifications

TEEP Scholar

Ministry of Education, Taiwan

Sep 2025 - Nov 2025

Selected for Taiwan Experience Education Program for research on predicting time-series voltage of MFC's using hybrid LSTM-GRU models under Prof. Wang Chin-Tsan.

SPARKLE Research Fellow

Let's Rewire EV Battery Charging

Oct 2023 - May 2025

Received INR 2 Lakh grant to pursue research on ML-based EV battery charging optimization. Led a 10-member team for experiments and model development.

PRMO Scholar

Pre-Regional Mathematics Olympiad

Oct 2018

One of 200 students from Delhi Region who qualified for Regional Mathematics Olympiad (RMO).

Man of the Match

Cult Sports Fest 2024

Apr 2024

Won the award for scoring a hat trick in the Rocketium football tournament.

Certifications

AI and Edge Computing

L&T Edutech - Apr 2024

Intermediate Machine Learning

Kaggle - Aug 2022

Intro to Machine Learning

Kaggle - Aug 2022

Get in Touch

Let's Build Something Amazing Together

I'm always open to discussing new opportunities, interesting projects, or potential collaborations. Whether you have a question or just want to say hi, feel free to reach out!

Send Me an Email Download Resume