Software Engineer / AI Engineer /ML Engineer

Rishab Darshan Shylendra

**********@*****.*** LinkedIn GitHub +1-716-***-****

TECHNICAL SKILLS

• Languages: Python, Java, C, R, Kotlin, JavaScript, TypeScript, SQL, Go, Rust

• Machine Learning/GenAI: PyTorch, NumPy, Pandas, NLP, Agentic AI, LLMs, RAG, LangChain, MLOps, OpenCV, GenAI, LlamaIndex, Spark, Kafka, PySpark, Reinforcement Learning, Diffusion models, JAX, speech-to-text (STT), text-to-speech (TTS)

• Frameworks: Django, Flask, Spring Boot, Angular, React/Next.js, REST APIs, FastAPI, Redis, Celery

• Data/Cloud: PostgreSQL, SQLite, NoSQL, MongoDB, Docker, AWS, Git, GitHub Actions, Linux, Jira, Google Cloud, Kubernetes PROFESSIONAL EXPERIENCE

University at Buffalo, SUNY Buffalo, NY

Research Assistant (Advisor: Dr. Alina Vereshchaka) Feb. 2025 – Present

• Designed a Self-Supervised Learning (SSL) framework to benchmark contrastive methods (SimCLR, MoCo v3, BYOL) for cross-domain brain disease classification on unlabeled multi-site MRI data; published and presented the research findings at the IEEE CogMI 2025.

• Developing a RAG and LLM-based Explainable Image Retrieval system for clinical knowledge graph construction and hierarchical MRI data reasoning, targeting a 60% reduction in manual radiology reporting through automated, interpretable medical report generation. Tata Consultancy Services (TCS) Bengaluru, India

Software Engineer Dec. 2021 – Jul. 2024

• Engineered Android front-end and Spring Boot/Tomcat backend for BSNL 5G telecommunications project, improving hotspot connectivity by 30%; led a 3-member team to automate and deploy Angular based field-testing GUI, boosting client satisfaction by 15%.

• Implemented RESTful API features across ANDSF and 3GPP protocols for the BSNL project, integrating 4+ microservices; engineered Appium-based automation scripts for multi-device testing across Android versions, reducing bug resolution time by 40% and integration defects by 25%.

• Optimized an ML recommendation engine (XGBoost) for the SonyLIV streaming platform through hyperparameter tuning, feature engineering, and cross-validation, achieving a 15% increase in user engagement through personalized content recommendations and improved model precision.

• Mentored an 8-member induction team on Java, Android, Spring Boot, and Agile development practices, achieving 100% on-time onboarding; earned ‘Special Initiative’ and ‘Xcelerate Warrior’ awards for demonstrating technical excellence, teamwork, and leadership. RESEARCH PAPERS

Self-Supervised MRI Representation Learning for Cross-Domain Brain Diseases Classification Feb. 2025 – Jul. 2025 Published – IEEE CogMI 2025 (Advisor: Dr. Alina Vereshchaka)

• Pre-trained and fine-tuned SimCLR, MoCo v3, and BYOL across four multi-site brain MRI datasets, achieving 84% classification accuracy with minimal labeled data; integrated Grad-CAM for interpretable, class-discriminative predictions across four brain diseases. PROJECTS

Wheniverse March. 2026 – Present

Python, Django, Tavily API, LLM, Web Search, MongoDB

• Developing an AI full-stack event notification platform tracking 100+ topics from live sources with 24/7 monitoring to deliver concise, real-time alerts; built using Tavily API scraping, Django backend, and LLM-powered summarization for contextual event discovery.

• Engineered an LLM-powered natural language query interface with multi-source web data aggregation and contextual ranking; built a responsive frontend enabling semantic event search and personalized feeds, achieving a 10% increase in user sign-ups at an hackathon. PaperDigest AI Feb. 2026

Next.js, Django, Celery, Redis, PostgreSQL, RAG, OCR, Gemini, Ollama

• Architected a full-stack AI research assistant using Next.js and Django with a custom RAG pipeline to parse uploaded academic PDFs, identify research gaps, and synthesize well-structured summaries from the literature; improved section-wise summary coherence by 40%.

• Designed a high-concurrency infrastructure with asynchronous background processing via Celery and Redis, supporting concurrent multi- user document analysis sessions with sub-second response times, scalability, and zero performance degradation under heavy load. Noetica Studio (Photo Editor) May. 2025

Computer Vision, JavaScript, HTML5 Canvas, CSS

• Constructed a browser-based photo editor with real-time preview using HTML5 Canvas API, supporting crop, rotate, brightness/contrast, and custom filter effects; optimized rendering for large images across modern browsers, improving efficiency by 35% for complex assets.

• Implemented a layered undo/redo history system, keyboard shortcuts, and one-click export to PNG, JPEG, and WebP formats, enabling a fully non-destructive editing workflow in the browser with zero server-side overhead and sub-millisecond operation latency. EDUCATION

University at Buffalo, SUNY Buffalo, NY

M.S. in Computer Science and Engineering (GPA: 3.8/4.0) Jan. 2026 Global Academy of Technology Bengaluru, India

B.E. in Computer Science and Engineering Aug. 2021

Contact this candidate