Hi, I am

Deepayan Sur

Machine Learning Engineer

I am passionate about leveraging data to drive innovation, I specialize in data science and machine learning, crafting impactful solutions for real-world challenges. With expertise in AI, data analysis, and software engineering, I’m dedicated to building data-driven systems that shape the future.

As a May 2025 graduate from Viterbi School of Engineering, I am actively seeking Data Science and Machine Learning Engineering roles, with additional interest in Software Engineering and Data Analysis.

I am currently Open to Full-Time Opportunities in 2025.

About Me

I am a Computer Science Master’s student at USC, specializing in Machine Learning, NLP, and AI-driven systems. My expertise lies in building scalable, data-driven solutions, designing predictive models, and developing NLP systems that drive efficiency and automation.

With experience in large-scale data processing, model deployment, and AI research, I have worked on user behavior analytics frameworks, enterprise NLP solutions, and high-impact research in Retrieval-Augmented Generation (RAG) and logical reasoning models. My work has led to a 70% boost in efficiency for data-driven systems, 30% improvements in AI-driven formula generation, and mentorship for 150+ trainees in data engineering and ML workflows.

Currently, I am actively working on AI-driven synthesis engines and scalable ML pipelines, refining LLM models for structured reasoning, and leveraging deep learning to enhance interpretability in scene-based AI models.

I am actively seeking full-time roles in Data Science and Machine Learning Engineering for 2025, bringing expertise in AI, data pipelines, NLP, and large-scale analytics to solve real-world challenges.

🚀 Technologies & Tools I Work With:
  • Programming Languages: Python, SQL, C++, Java
  • Machine Learning & AI: Scikit-learn, TensorFlow, PyTorch, JAX, LangChain, Hugging Face
  • Data Processing & Engineering: Pandas, NumPy, Spark, MLflow, DuckDB, Apache Airflow
  • NLP & Generative AI: Spacy, Transformers, LlamaIndex, LoRA
  • Cloud & DevOps: AWS, Docker, Kubernetes, Weights & Biases
  • Model Deployment: FastAPI, Flask, Triton Inference Server

Experience

Graduate Researcher - USC CPS-VIDA Lab
April 2024 – Present
  • Developed an AI-driven synthesis engine, improving user interaction through intelligent recommendation methodologies and managing 14+ concurrent distributed software systems using decision trees.
  • Engineered a Retrieval-Augmented Generation (RAG) framework with LangChain, achieving a 30% improvement in TQTL formula generation accuracy for translating natural language to logical representations.
  • Fine-tuned LLM models via Transfer Learning and RLHF, boosting TQTL prediction precision by 25% on complex natural language datasets for scene interpretation.
  • Accepted two research papers at SAC 2025 and VMCAI 2025.
Machine Learning Engineer - Highradius Technologies
July 2022 – June 2023
  • Designed and developed a scalable framework for user behavior analysis on 1.2M+ user patterns, refining recommendation algorithms for Autonomous Collections, improving user efficiency by 70%.
  • Implemented an Email Intent Classifier and Email Entities Recognition system using Spacy and Named Entity Recognition (NER), saving 10% of collector bandwidth daily.
  • Led training sessions for 150+ trainees and mentored 12 interns, collaborating cross-functionally to implement data pipelines for downstream ML tasks.
Data Science Intern - Highradius Technologies
April 2021 – June 2022
  • Engineered an end-to-end Cash Flow Predictor for 4+ client accounts, focusing on data preprocessing, analysis, and time-series forecasting.
  • Automated Payment Posting Software across 3+ ERPs, integrating API-based automation for improved efficiency.

Education

Aug 2023 – May 2025
Master of Science in Computer Science
University of Southern California
GPA: 3.57 out of 4.0

Relevant Coursework:

  • Machine Learning, Natural Language Processing, Deep Learning
  • Advanced Algorithms, Database Management,Robot Learning
Jun 2018 – May 2022
Bachelor of Technology in Computer Science Engineering
Kalinga Institute of Industrial Technology
GPA: 9.21 out of 10.0

Relevant Coursework:

  • Algorithms, Database Management, Software Engineering
  • Computer Networks, Operating Systems, Automata & Formal Languages

Projects

Language Detection in Low Resource Code-Switched Text
NLP BiLSTM FastText
Language Detection in Low Resource Code-Switched Text
Developed a BiLSTM-based model achieving 96.31% accuracy in Nepali-English detection and 0.81 F1 score in Hindi-English sentiment analysis.
CaptionCraft: AI-Powered Image Captioning
Computer Vision Transformers CNN
CaptionCraft: AI-Powered Image Captioning
Built an AI-powered captioning system using Computer Vision Transformers and NLP, increasing social media engagement by 17.6%.

Get in Touch

My inbox is always open. Whether you have a question or just want to say hi, I’ll try my best to get back to you!