Aryan Raj
Machine Learning Engineer & Backend Developer
🎓 Education
SRM Institute of Science and Technology
Chennai, INB.Tech Computer Science and Engineering with spl. in Artificial Intelligence and Machine Learning
May 2021 - May 2025
CGPA: 8.4/10
💼 Experience
Machine Learning Engineer
Hyderabad, INSEOstackSep 2025 - Present
- • Building and deploying autonomous AI agents to automate complex SEO workflows and content analysis pipelines, reducing manual effort by 60% and improving turnaround times.
- • Designing and implementing scalable machine learning infrastructure using LangChain, LangGraph, and LlamaIndex to support multi-step reasoning and real-time content optimization.
- • Architecting prompt engineering frameworks and evaluation systems to ensure high-quality, contextually relevant SEO recommendations across diverse client domains.
Machine Learning Engineer (Intern)
Bengaluru, INHyperVergeAug 2024 - July 2025
- • Developed and fine-tuned LLM-based solutions to automate KYC and fraud detection workflows, tailored for real-world, domain-specific regulatory use cases across financial institutions.
- • Led evaluation and benchmarking of LLMs on over a million real-world data points, setting up scalable performance monitoring pipelines and achieving industry-accepted False Acceptance Rate (FAR) and False Rejection Rate (FRR) thresholds.
- • Optimized and deployed state-of-the-art computer vision and NLP models for identity verification, significantly improving inference speed and accuracy under production constraints.
Machine Learning Intern
California, USA (Remote)Emendo AIJan 2024 - Aug 2024
- • Engineered and deployed AI solutions leveraging the AWS ecosystem with a focus on Generative AI to build scalable, client-centric applications.
- • Developed and maintained 10+ Generative AI-based microservices with industry-standard integrations using AWS Lambda, API Gateway, and OpenTelemetry for observability.
- • Architected scalable, serverless backends to support efficient retrieval and generation workflows across diverse application domains.
Research Intern
Chennai, INIndian Institute of Technology, Madras (IIT M)Feb 2023 - Jul 2023
- • Collaborated with the Department of Ocean Engineering to design and implement a machine learning-based anti-collision system, improving accuracy of existing solutions by over 28%.
- • Developed marine object detection, tracking, and localisation systems using stereo vision-based camera setups for alternative navigation in unmanned surface vehicles (USVs).
- • Worked with state-of-the-art computer vision models and successfully deployed the solution on edge-based IoT hardware for real-time maritime applications.
🚀 Projects
PaperPilot
Next.js, TailwindCSS, Python, Flask, Pinecone, Jina AI, LangChain
- • Built a full-stack AI-powered research assistant platform for semantic search and summarization of academic papers with high precision retrieval.
- • Implemented advanced RAG pipelines using Pinecone for vector storage and Jina AI embeddings, enabling context-aware paper discovery.
- • Developed a Next.js frontend with TailwindCSS and a Flask backend, processing over 500+ papers with structured metadata extraction and citation analysis.
Open-KYC ⭐
Next.js, OpenCV, TensorFlow, Tesseract, ShadCN, WebRTC
- • Built an AI-based KYC portal with features like facial authentication, Aadhaar/PAN card OCR, and liveness detection via video stream.
- • Used OpenCV and TensorFlow for identity verification workflows, achieving high accuracy and security standards.
- • Secured 1st place at Standard Chartered Hackathon 2024 for end-to-end automation and production readiness.
Educative.AI
Python, FastAPI, React.js, TailwindCSS, LLMs, TensorFlow
- • Developed a student assistance tool integrating OCR and 10+ fine-tuned open-source LLMs to process handwritten and blackboard notes.
- • Implemented a FastAPI backend and React.js frontend for features such as MCQ generation, speech-to-text doubt resolution, and structured note organization.
- • Enabled automatic resource retrieval and summarization, enhancing accessibility and productivity for students.
AI-RoadGuard 🏆
React.js, CNN, Flask, Python, TensorFlow
- • Engineered a CNN-based accident detection system achieving over 90% accuracy with 50% fewer false alerts than existing solutions.
- • Developed a Flask backend and React.js frontend dashboard for real-time alerting to emergency services.
- • Recognized as the Best Project in Open Innovation at MLH MesoHack 2022.
OneMed 🏆
Python, Next.js, JavaScript, LLMs, Pinecone, MongoDB, AWS
- • Built a full-stack AI-powered EHR platform for hospitals, integrating LLM-based summarization and voice-to-text consultations.
- • Developed a React Native app for syncing emergency data and patient conditions in real-time.
- • Implemented scalable vector search with Pinecone for intelligent patient record retrieval; ranked Top 10 at MozoHack 2024.
⚡ Technical Skills
Languages
CC++PythonJavaJavaScriptSQL (Postgres)HTMLCSSR
Frameworks
React.jsNext.jsNode.jsFlaskFastAPITailwindCSSJUnitMaterial-UI
Machine Learning
NumPyPandasScikit-learnTensorFlowPyTorchOpenCVNLPLangChainLangGraphLlama-IndexTransformersHuggingFace
Developer Tools
GitDockerKubernetesAWSAzureGoogle Cloud PlatformCDKTerraformRedisPinecone
📚 Publications
Structured Relevance Assessment for Robust Retrieval-Augmented Language Models
Aryan Raj, Astitva Veer Garg, Dr. Anitha D
Conference: International Conference on Information Technology for Social Development (ITC4SD 2025)
Publisher: Springer Nature
This paper explores the application of fine-tuned LLMs for automated evaluation of open-ended questions in educational contexts, demonstrating significant improvements in accuracy and consistency over traditional methods.
🏆 Achievements
- Secured 1st place at the Standard Chartered Hackathon for developing OpenKYC, an innovative KYC solution that streamlined identity verification processes.
- Won 2nd place at Hack Nova 2024 with Educative.AI, an educational technology project later selected to represent at Innverve 2023, Army Institute of Technology (AIT), Pune.
- Received the Best Project Award in the Open Innovation category at MLH Meso Hack 2022 for the project AI-Roadguard.
- Authored technical articles on AI for prestigious Medium journals, including the DataX Journal.
👥 Club and Contributions
- Contributed to various open-source projects; notable contribution includes Dify, a project with over 100,000 stars on GitHub.
- Conducted research at Next Tech Lab (2021-2025) as a member of Norman and McCarthy Labs, collaborating on web and machine learning projects, specializing in deep learning for image-related tasks.
- Served as Technical Director at Data Science Community SRM (2022-2023), organizing technical events, workshops, and hackathons including DS Hack 2.0.
- Led machine learning initiatives at SRM Quantum Computing Club, managing projects at the intersection of quantum computing and machine learning.
- Served as Event Domain Member of SRMKzilla, the official Mozilla club on campus, promoting community contributions and engagement.
Get In Touch
Have a project in mind or want to collaborate? I'd love to hear from you!