🌟 Featured Initiative: AI From Scratch
🧠 aifromscratch.in — The Ultimate 30-Day Guide to the Mechanics, Math, and Architecture of Artificial Intelligence.
A curated, interactive curriculum demystifying everything from Tokenization, Attention, and Transformers to Hybrid RAG and Autonomous Agents.
👉 Explore the 30-Day Roadmap & Paper Breakdowns →
I am an AI and Data Engineer bridging the gap between rigorous academic NLP research and enterprise-scale production systems.
- 🏦 JPMorgan Chase & Co.: Architecting fiduciary-grade, hybrid Retrieval-Augmented Generation (RAG) pipelines integrating sparse keyword and dense semantic vector search for institutional financial intelligence.
- 👟 Ex-Nike: Engineered high-throughput distributed ETL pipelines, ML feature stores, and automated data workflows serving global e-commerce workloads.
- 🎓 IIT Gandhinagar: Completed M.Tech in Computer Science with a 9.67 CGPA, specializing in NLP, Machine Learning, and Information Extraction.
- LEGOBench: Scientific Leaderboard Generation Benchmark
Findings of the Association for Computational Linguistics: EMNLP 2024
A unified benchmark evaluating LLMs and scientific encoder architectures across 22 years of arXiv preprint data and 11,000+ PapersWithCode leaderboards.
[Paper (arXiv)] • [Codebase] • [Dataset (OSF)]
| Domain | Stack & Frameworks |
|---|---|
| AI / NLP | PyTorch, HuggingFace Transformers, LangChain, LlamaIndex, Vector DBs (Chroma/Milvus/Pinecone), Fine-tuning |
| Big Data & Pipelines | Apache Spark, Apache Airflow, Kafka, Jenkins, ETL Pipeline Design |
| Full Stack & Backend | FastAPI, Node.js, Next.js, React, GraphQL, REST APIs |
| DevOps & Cloud | Docker, Git, CI/CD Actions, Linux/Bash, Selenium |
| Project | Description | Tech Stack |
|---|---|---|
| 🧠 AI From Scratch | Interactive 30-day AI curriculum and learning platform covering foundational math to production agents. | Next.js, React, TypeScript, KaTeX |
| 🛡️ git-secret-scanner | Automated static analysis tool to detect and prevent committed credentials in Git history. | TypeScript, Node.js |
| 🔍 AI-Image-Search | High-speed semantic visual search engine powered by multi-modal deep embeddings. | Python, PyTorch |
| 📊 Table-Mapper-AI | Document intelligence pipeline to identify, extract, and structure irregular tabular data. | Python, Computer Vision |
| 📖 Al-Quran | Computational linguistics analyzer and semantic verse retrieval system. | Python, NLP |
| 🤖 auto-quora-answer-bot | Context-aware automation pipeline leveraging LLMs and browser automation. | Python, OpenAI, Selenium |
💬 Let's collaborate on production AI systems, LLM evaluation, and high-performance data engineering.
Designed with care by Shoaib Alam