My Experience
Software Engineer Intern
June 2025 - PRESENT | Los Angeles, CA
@Grad8- Engineered an AI-powered grading platform using multimodal LLMs (GPT-4o, Claude 3.7, Qwen-VL) + OCR to score handwritten/text submissions; added batch grading, multithreading, and secure file ingestion with AWS RDS + FastAPI for high-throughput processing.
- Developed rubric-generation dashboards and deployed a production backend on AWS serving 150+ instructors and 10K+ grading requests/day, cutting grading turnaround time by 60%.
ML Research Assistant
January 2025 - PRESENT | Los Angeles, CA
@USC AD Labs- Implemented a GPT-style Decision Transformer for offline RL in autonomous driving—leveraged return-to-go, state, and action embeddings for long-horizon planning without value-bootstrapping and built end-to-end pipelines from trajectory logs to HD map vectors with discretized action spaces.
- Architected a spatial-temporal encoder merging HD-map geometry (road layout, lane direction) with ego-agent context (past trajectories) via multi-head attention in closed-loop planning performance and exploring LLMs.
Software Developer Engineer
August 2023 - July 2024 | Ahemdabad, Gujarat, India
@Aasma Technology Solutions- Formulated cross-platform data-sharing software with microservice architecture; built pipelines in Jenkins (Blue Ocean) to automate VM testing, executables, and integration/unit tests (PyTest, Mockito), working with Microsoft/GitHub team in Agile Scrum meetings.
- Troubleshot and optimized large-scale API integrations with OOP, AWS Glue, and S3, improving response times by 30%. Led team of 10 for ETL workflows, boosting pipeline efficiency 68% and cutting technical debt 20%.
Full Stack Develoepr Inter
May 2023 – August 2023 | Ahmedabad, Gujarat, India
@Aasma Technology Solutions- Developed a MERN-based resource management platform (CRM) with WebSockets, RESTful APIs, and a microservice architecture, addressing large-scale challenges in code review syncing 2M+ records across locations.
- Deployed Azure SQL with Jenkins for automated testing, leveraging Google Analytics to reduce resource wastage by 42%, increase lead conversions by 15%, and expand market reach.
Data Analyst - Research Fellowship
May 2023- August 2023 | Ahmedabad, Gujarat, India
@PDEU- Designed and implemented a malicious URL detection system on a dataset of 650K+ URLs, applying advanced statistical feature engineering and machine learning models (Gaussian NB, Random Forest) alongside deep learning architectures (RNN, LSTM, Bi-LSTM, GRU, CNN).
- Achieved 96%+ detection accuracy, building a feature-rich pipeline with 21 behavioral indicators (e.g., IP presence, URL length, suspicious keywords) and generating actionable insights into phishing and malware trends.
Web Developer Intern
November 2022 - January 2023 | Surat, Gujarat, India
@LST Capitals- Engineered cross-platform (MacOS & AndroidOS) data-pipeline framework using Databricks, Flask API, and Firebase to enable real-time analytics on 50K+ records with automated testing and integrated Web3 support.
- Integrated a full-stack, PCI-compliant payment infrastructure supporting multi-currency cards, UPI, wallets, and PayPal with real-time analytics, refund/dispute automation, fraud detection, and webhook-driven.
My Skills
Languages:
Web Tech:
AI & ML:
Database & Cloud Platforms:
Tools & Methodologies:
A small selection of recent projects
PixMod
AI-powered SaaS platform using SAM for image segmentation (object removal, recoloring, background removal) and Stable Diffusion for generative tasks like restoration and inpainting. Integrated CLIP enables visual element-based image search, with Stripe for payments.
View
Big Brain
Built an AI-powered document platform with Pinecone, FAISS, Convex, and Clerk, enabling secure uploads, intelligent analysis, real-time summaries, fast vector search, and collaborative discussions in a Shadcn UI.
View
HIMS - Hospital Management System
Gujarat State Hospital Management System for managing patient records, billing, scheduling appointments and other hospital management tasks.
View
FinNexus - Personalized Banking System
Personalized banking system for managing accounts, transactions of debit and credit cards, loans, and other banking tasks.
View
CPU Scheduler
Operating System project for simulating CPU scheduling algorithms like FCFS, SJF, SRTF, RR, Priority and Multilevel Queue.
View
Quantum Resistant Blockchain Technology
Proposed a quantum-resistant blockchain framework using FALCON digital signatures and the DPoSB consensus algorithm. DPoSB ensures decentralization by monitoring node behavior and using Borda Count for fair witness node selection, maintaining security and transparency.
View
My approach
Requirement Gathering & Planning
I start by gathering detailed requirements from stakeholders to fully understand their needs and expectations. I then define the project’s scope, objectives, and constraints to set clear goals. After that, I create a comprehensive project plan that outlines timelines, resources, and budget. This plan acts as my roadmap, ensuring that I stay on track and align with the project's vision.
Development & Progress Update
In this phase, I design the system’s architecture and components based on the requirements I’ve gathered. I create detailed specifications, wireframes, and prototypes to guide the development process. I then proceed with coding and developing the system according to these designs. Throughout development, I perform continuous testing to identify and address issues early, making sure the system meets all requirements and quality standards.
Development & Launch
Once development is complete, I deploy the system to the production environment. This involves configuring the environment, migrating data if needed, and ensuring all components are properly integrated. After deployment, I monitor the system’s performance and provide ongoing support to address any issues. I also gather user feedback to make necessary improvements and updates, ensuring the system remains effective and up-to-date.