Suyash Dubey
SDE | ECE @ IIITD | C4GT'23 @ NHA | MERN stack Developer | ML enthusiast | Contributor at GSSOC'23
New Delhi
3+ years of work experience
Oct 20261 moPresent
Gurgaon
Oct 20258 moJun 2026
Remote
May 20255 moOct 2025
Gurgaon
Mar 20252 moMay 2025
Remote
May 202410 moMar 2025
Remote
Project: Developing LLM interfaces using AWS bedrock (Detailed demo: https://suyashblog.hashnode.dev/llm-app-dev-using-aws-bedrock-and-langchain)
Methodologies: created an LLM-powered question-answering (QA) application over large document corpora using AWS Bedrock and LangChain. I handled challenges like managing large documents exceeding token limits and finding relevant documents. The process involves loading and processing documents, creating vector embeddings with Amazon Titan, and storing them in a vector database like SingleStore. When a question is asked to the LLM, relevant document chunks are retrieved and used as context for generating answers with models like Anthropic Claude through Amazon Bedrock, following a Retrieval-Augmented Generation (RAG) approach.
Tech Stack: AWS Bedrock, AWS titan, SinglestoreDB, LLMs, RAG
Oct 20232 mosNov 2023
Remote
Experimenting with different LLMs to design a new LLM for tutoring students of all ages.
Doing prompt engineering to train an innovative AI model by providing human responses to different questions. Engaging in research to explore the latest advancements and push the boundaries of what AI can create. Worked to address issues like bias mitigation and responsible AI use, ensuring our creations adhere to ethical standards.
Aug 202310 mosMay 2024
New Delhi, Delhi, India
Evaluation of the performance of hardware accelerators for implementation of Neural Networks using in-memory computation
Advisor: Dr. Anuj Grover
-828dbcbe.jpg&w=3840&q=75)
Jul 20234 mosOct 2023
- Worked in a team of 4 developers along with a mentor on the backend framework of the Digital health portal of the National Health Authority. Specifically, working on synchronizing asynchronous API workflows by creating efficient, user-friendly interfaces and facades.
- Implemented the feature of Subscription management flow in ABHA app, which helped automate Subscription Requests. I emphasized developing and utilizing microservices to create a more end-to-end and responsive system.
- Contributed to the UI of the app, ensuring a seamless user experience. We heavily leveraged SOA to break down complex functionalities into smaller, manageable services, enhancing the flexibility of our system. Using a NoSQL-based database like mongoDB to store a mapping of different services and troubleshooting any errors to ensure high-quality code.
- Testing the APIs with the interfaces built to provide a seamless experience to the users and deploying changes on ABHA app with the help of ABDM team. Helping ABDM developers to maintain the code and help in software improvement seamlessly.
- Tech stack: Springboot, Java, gradle, NoSQL, Docker, Ngrok
Jun 20232 mosJul 2023
Remote
Omdena projects address real-world challenges with team of 50 AI engineers who collaboratively develop real-world solutions. We developed a fully functional ML web app for Goe Wellness with appropriate ML models and algorithms. It involved data collection, analysis, Activity tracking, building appropriate recommendation system etc. Tech stack: Python, PyTorch, scikit, pandas, seaborn
Suyash’s AI Stack
Education
Bachelor's degree, Electrical, Electronics and Communications Engineering
Indraprastha Institute of Information Technology, Delhi
20204 yrs2024
High School Diploma
ISG jubail
Join Suyash and 10k+ other professionals on
Tal is the app where bosses meet engineers & builders directly, without middlemen
Download Tal Boss