Software Engineer

Aditya Thaker

**************@*****.*** 206-***-**** github.com/ADIthaker www.linkedin.com/in/aditya-thaker2811/ EDUCATION

Master of Science in Computer Science, University of Colorado Boulder ; GPA: 3.95 Aug 2023 – May 2025 Relevant Coursework: Design and Analysis of Algorithms, Compiler Construction, Distributed Systems, Natural Language Processing, Linux System Administration, Data Center Scale Computing. Bachelor of Engineering in Computer Engineering, University of Mumbai; GPA: 3.78 Aug 2019 – May 2023 TECHNICAL SKILLS

Languages/ Databases: C++, Python, JavaScript, Go, C, Java, MongoDB, PostgreSQL, MySQL, Redis, DynamoDB. Web Development: Node, React, Django, nginx, HTML, CSS, Bootstrap, REST API, GraphQL. Cloud/Big Data: GCP, AWS, GitHub, Spark, PySpark, Kafka, Ansible, Hadoop, Docker, Kubernetes. AI/ML: TensorFlow, pandas, scikit-learn, numpy, torch-geometric, OpenAI, Langchain, LangGraph. PROFESSIONAL EXPERIENCE

University of Colorado, Boulder, USA May 2024 – Jan 2025 Graduate Researcher

• Engineered a high-performance video conferencing WebRTC server in Go, resulting in 35% improvement in latency.

• Leveraged XDP for packet filtering, achieving 50% reduction in processing time for incoming packets.

• Implemented traffic routing, enabling dynamic streams for users, using eBPF Maps on Linux kernels. SimPPL, Mumbai, India Jun 2022 - Jun 2023

Fullstack Web Developer

• Optimized analytics visualizations, using virtual DOM updates and Canvas rendering, reducing load times by 25%.

• Built an insightful analytics dashboard using d3.js and plotly to track user engagement with news articles, integrating real-time trend analysis.

• Enhanced the REST API with Django, and optimized PostgreSQL schemas, improving response times by 28%. University of Mumbai, Mumbai, India Jun 2022 - Jun 2023 Backend Web Developer

• Led end-to-end development of a MERN web app to automate exam duty assignments for 100+ examiners across 8 departments, used Redis to perform CRUD, before persisting on MongoDB.

• Collaborated with stakeholders to define project scope, gather requirements, and ensure the solution aligned with department needs, fostering clear communication and teamwork.

• Configured and deployed application, setup payments with Stripe, and automated emails, reducing manual work. IIT Patna, Patna, India Nov 2021 – Jun 2022

ML Research Intern

• Designed a cryptocurrency analytics pipeline using Graph ML to detect fraudulent users in PyTorch.

• Devised a graph ML model to develop a phishing user classifier trained on Ethereum data, achieving 98% F-score.

• Conceptualized ego graphs to convert transactions into temporal transactions, accelerating training time by 20%. Mathrithms, Mumbai, India Jun 2021 - Aug 2021

Backend Development Intern

• Redesigned the PostgreSQL database for a high-traffic food ordering platform, implementing indexing to improve query performance by 30%.

• Developed an algorithmic stock trader deployed using nginx, implementing predefined strategies to fully automate client operations, doubling execution speed.

• Deployed automated CI/CD on AWS ensuring seamless updates, scalability, and high availability for clients. PROJECTS

Content Curation News Feed

• Built an LLM-powered news feed that curates personalized content using vector databases for preference storage.

• Optimized feed generation with Kubernetes, Kafka, and real-time embeddings, cutting load times by 45%.

• Integrated observability with Prometheus and Grafana, enabling proactive tuning and improved reliability. Win An Argument

• Designed a multi-tool LangChain AI agent that intelligently selects between research retrieval, summarization, and fact-checking tools, improving response accuracy.

• Designed an efficient retrieval system, enabling real-time search across 100K+ research articles with citation accuracy.

Emotion Cause Analysis in Conversations

• Designed a dialogue processing pipelines on separate cloud instances and improved architecture to increase throughput by 20%.

• Tested and evaluated model architectures like BERT, CRFs, RoBERTa and achieved a maximum F-score of 85%.

Contact this candidate