I’m an AI/ML Engineer & Data Scientist with hands-on experience building predictive models, end-to-end ML pipelines, and LLM-powered applications. My work spans across industry, research, and hackathons — centered on using AI to drive real-world impact. At DF Young, I’m currently leading efforts to extract hidden business value from logistics data for enterprise clients. From developing predictive reorder models to prototyping Graph Neural Networks (GNNs) for uncovering complex relationships, I focus on delivering actionable insights that improve supply chain decision-making. At DSM-Firmenich, I built machine learning and deep learning models that enhanced flavor formulation accuracy by 20%.

I'm passionate about leveraging cutting-edge AI, including Generative AI, LLMs, and Graph Neural Networks, to solve complex business problems, particularly in supply chain and logistics. I am eager to connect with engineers, researchers, and founders who are building innovative, data-driven products. I'm currently seeking full-time AI/ML Engineer or Data Scientist roles where I can contribute to developing impactful, end-to-end machine learning solutions.

Hi, I’m Jeet Choksi, Software Engineer & Data Scientist with 3+ years building low-latency microservices, ML inference systems, and cloud-native data pipelines. I earned my M.S. in Data Science from CU Boulder (GPA 3.7) and ship production systems on AWS/GCP using Python/C++/PySpark/Kubernetes. I’ve led teams, won hackathons, and delivered measurable wins p95 latency <100 ms, 1k+ req/s services, and 40% faster ETL. I care about elegant engineering, observability, and turning data into impact.

High-performance systems (C++/Rust, lock-free/multithreaded design), ultra-low-latency microservices, and streaming data (Kafka/Flink). LLM systems & RAG (retrievers, rerankers, eval harnesses), vector search, and cost/latency optimization. Production MLOps (Kubernetes, feature stores, monitoring). Open to collaborating on infra for real-time ML, self-optimizing retrieval layers, and AI agents for ops/DevEx.

Virtual Coach: real-time pose-feedback (FastAPI, MediaPipe/OpenCV, CUDA) with <100 ms latency on GKE; adding rep-counting + form-score model. • Vaani News: LLM/RAG summarization + Hindi TTS, Redis/Celery pipelines on GCP; experimenting with on-device distillation. • RecomAI: BERTTopic+MongoDB recommender with Kafka ingestion; building online-learning + bandit exploration. • Self-optimizing RAG layer: auto-tunes chunking/embeddings/rerankers via live metrics; Airflow jobs reindex nightly.

Deepesh Haresh Katudia is an aspiring software developer with one year of experience, currently serving as a Software Developer Intern at Get Superstar Inc. He pursues software development with an AI focus, reflected by his tinkerer role as Software Developer (AI Focus), and is open to work, seeking full-time opportunities. Deepesh combines hands-on internship experience with early-career ambition to transition into a full-time software development role; his LinkedIn is https://www.linkedin.com/in/deepesh-katudia-0a46a7207/ and he uses a gmail.com email address for contact.

Full-stack web development, Applied machine learning, TypeScript/Next.js, React, Node.js/Express, Python, TensorFlow/CNN, OCR/computer vision, OpenAI API integration, Tailwind CSS, Vercel deployments, Git

Online-Job-Platform — full‑stack TypeScript/Next.js app with React frontend and server-side APIs; Portfolio website — Next.js site deployed to Vercel; CodeMentor_AI — React frontend + Node/Express backend integrating OpenAI API for mentor/QA features; Indian Currency Detection — Python notebooks with CNN/TensorFlow models and OCR pipelines for currency recognition and detection.

I am Deep Mehta, a data scientist and AI researcher passionate about building machine learning, natural language processing (NLP), and computer vision solutions. I am currently pursuing a Master’s in Data Science and actively developing AI-driven projects, including real-time chatbots, AI agents, and intelligent automation tools. I specialize in Python, SQL, LangChain, OpenAI models, and vector databases like ChromaDB, using these technologies to create web content interaction tools, skill development trackers, and automated job application systems. I enjoy participating in hackathons, collaborating with industry experts, and pushing the boundaries of AI innovation. I continuously explore retrieval-augmented generation (RAG), autonomous agents, and real-time AI processing to develop transformative AI solutions. My goal is to enhance efficiency, decision-making, and automation across industries through cutting-edge AI applications.

I am currently interested in learning more about advanced machine learning techniques, especially in the areas of natural language processing (NLP) and computer vision. I'm particularly drawn to the application of deep learning models like transformers for NLP and object detection for computer vision tasks. I'm also exploring AI ethics, fairness, and explainability in machine learning models. In terms of potential collaborators, I'm looking to connect with professionals or researchers who specialize in: AI and ML research, particularly in developing cutting-edge models Data science and analytics, with a focus on solving real-world business problems Software engineers with expertise in deploying machine learning models to production AI ethics and policy experts, especially those with experience in ensuring fairness, transparency, and accountability…

I’m currently working on building an AI-powered chatbot that interacts with websites in real-time, extracting and analyzing relevant content dynamically. The project leverages LangChain, OpenAI embeddings, and ChromaDB to enable contextual responses based on website content. Additionally, I’m exploring AI agent development for various applications, including automated job applications (Job Bot) and document-based Q&A systems that help users interact with educational materials more effectively. Given my experience, retrieval-augmented generation (RAG), and real-time AI processing, I’m excited to push the boundaries of real-time voice and reasoning AI at this hackathon, integrating cutting-edge architectures for sub-second interactions.