About Me

Niharika Patil

Stressed, blessed, and coffee obsessed ☕

Hey, I'm Nicks — a software engineer who happens to have two ML degrees but would rather build a sick website than train another model. I love turning ideas into real, working products — whether that's a real-time chat app, a risk analysis platform, or a pastry shop website (yes, really). I'm the kind of person who'll learn a whole new tech stack over the weekend just to ship something cool. Currently looking for my next adventure in the UK.

I don't just write code — I design the experience, stress about the spacing, then rewrite the whole thing at 2am because I thought of a better way.

Experience

May 2025 - Sept 2025

AI Researcher

University of Nottingham

  • Developed the first systematic AI framework combining diffusion models with evolutionary insights for thermostable RuBisCO variants
  • Curated dataset of 36,000+ annotated sequences and 17,000 extremophilic proteins for multi-task learning
  • Achieved TM-scores of 0.91-0.98 validated via AlphaFold with preserved catalytic activity at 40-60°C

Dec 2022 - Feb 2023

ML Intern

AWS Academy, Pune, India

  • Built AWS Lex chatbot handling 10+ conversation intents with sub-2 second response times
  • Developed SageMaker ML models achieving 85%+ accuracy for mental health sentiment analysis
  • Integrated AWS Rekognition for image processing, improving workflow efficiency by 25%

Projects

Movie Mate

Movie Mate

Feb 2026 – Mar 2026

Multiplayer movie decision app sustaining 51 req/sec across 10 concurrent users with 0 errors. 13-endpoint PostgreSQL REST API with sub-400ms latency, 20 Jest/Supertest integration tests, containerised with Docker.

InvestX

InvestX

Nov 2025 - Jan 2026

Quantitative risk analysis platform processing 61 UK equities with 6 risk models including Monte Carlo simulation, VaR, and EWMA forecasting. Built with FastAPI achieving 7s response time.

Autonomous Driving

Autonomous Driving

Mar 2025 - May 2025

CNN-based steering model trained on 17,647 images achieving 35.8% improvement in prediction accuracy. Deployed real-time inference on Raspberry Pi with 180-200ms latency.

SMOTE on Spark

SMOTE on Spark

Mar 2025 - May 2025

Distributed SMOTE pipelines in PySpark handling extreme class imbalance (578:1 ratio). Achieved 91.49% balanced accuracy for fraud detection while reducing execution time by 49%.

BlinkChat

BlinkChat

Jun 2024 - Dec 2024

Real-time chat application with React and Firebase supporting 10+ concurrent users. Minimized latency to under 200ms and reduced bundle size by 35% using Vite.

DessertLove

DessertLove

Dec 2023 - Jun 2024

Responsive pastry shop website with online ordering, reservations, and blog features. Increased page load speed by 40% with lazy loading and optimized checkout by 25%.

Text-to-SQL RAG

Text-to-SQL RAG

Jan 2024 - May 2024

AI assistant translating natural language to SQL queries using LangChain and Mistral-7B. Implemented FAISS vector store achieving 95% accuracy on test set.

Skills

Programming Languages

PythonJavaScriptTypeScriptSQLRJavaHTML/CSS

Frontend & Frameworks

React.jsNode.jsViteTailwind CSSReact RouterElectron.jsFastAPIDjangoFlask

ML/AI & Data Science

TensorFlowPyTorchScikit-learnPandasNumPyOpenCVLangChainNLPDeep LearningCNNs

Cloud & Big Data

AWS SageMakerAWS LexAWS RekognitionFirebaseApache SparkPySparkDatabricksMongoDB

Education

2024 - 2025

Masters in Machine Learning in Science

University of Nottingham, United Kingdom

Merit (2:1) | GPA: 3.4/4.0

2020 - 2024

Bachelor of Engineering in AI & Machine Learning

Savitribai Phule Pune University, India

First-Class with Distinction | GPA: 3.53/4.0