Two years building intelligent systems that actually work in production. Robust NLP pipelines, real-time computer vision models, and secure cloud architectures on AWS. I care about shipping things that matter.
Engineer by training. AI developer by choice.
AWS Certified Software Engineer with two years of hands-on experience building robust AI/ML systems, automated Python pipelines, and scalable cloud-native architectures.
My work sits at the intersection of Natural Language Processing and Computer Vision. I write production pipeline workflows using HuggingFace, spaCy, OpenCV, and Tesseract OCR, orchestrating secure environments on AWS.
I love solving problems from scratch, whether that is training fine-tuned regressors for F1 telemetry or creating AI-powered agents to run DevOps troubleshooting loops.
Building deeper knowledge in LangGraph Multi-Agent Workflows, real-world Large Language Model Orchestration, and Gemini AI integrations through continuous self-study.
Building and optimizing production AI systems that deliver measurable business and cloud results.
Optimized key Python NLP workflows to accelerate document structure parsing speeds for massive legal text matrices.
Constructed high-speed API layers and intelligent document parsing pipelines combining Tesseract with OpenCV layout analysis.
Architected secure Amazon EC2/S3/Lambda cloud topologies, replacing manual integrity audits with automated ETL triggers.
Led legacy codebase audits to eliminate vulnerabilities, driving sprint targets 10% ahead of timeline schedule.
Hover over any card to view detailed competencies.
Click the "Live Demo" button on any card to run an interactive pipeline widget directly in your browser.
Multi-agent compliance auditor using LangGraph to evaluate LLM outputs against global standards. Employs fan-out/fan-in parallel checks, adversarial probing, and AI Compliance Passport PDF generation.
Stateful RAG-driven career mentor that maps technical resumes, indexes skills clusters inside FAISS Vector DB embeddings, and plots personalized upskilling paths via Llama-3 agent pipelines.
Intelligent, agent-driven monitoring pipeline that scans active container logs, detects CPU/memory anomalies, runs deep root cause analysis, and triggers automated self-healing pod restarts.
Data ingestion engine that processes Excel/CSV structures, cleans duplicate noise, draws rich business visualizations via canvas rendering, and answers natural-language business requests via the Gemini API.
Enterprise-grade serverless redaction pipeline that processes document scans with Tesseract OCR & OpenCV, classifies entities via HuggingFace Transformers, and stores redacted datasets on AWS S3.
Predicts options pricing models using Support Vector Regression (SVR), analyzes news sentiment via FinBERT models, and processes financial parameters using real-time yfinance APIs.
A high-fidelity document redaction engine that automatically censors sensitive PII (Names, Emails, Locations, Secret Coordinates) inside leaked transcripts, calculating a real-time Source Protection safety score.
High-speed server logs parser that runs regular expression pre-processing, performs anomaly classification using LangChain chains, and triggers real-time Slack incident notifications.
Predicts Formula 1 podium finishes using historical Ergast qualifying profiles, driver form indices, and live Open-Meteo weather parameters. Includes an optimal tyre compound strategy calculator.
Deep-dives and frameworks shared with the developer community. Click the tabs to switch articles, and explore the interactive visualizers on the left!
A comprehensive guide covering Retrieval-Augmented Generation from first principles through production architecture — including 9 RAG patterns, 8 search methods, vector database selection, fine-tuning workflows, and core LLM concepts like attention, MoE, and RLHF.
Have a role, project idea, or just want to chat? Reach out through any portal below.
Opens your system mail client with precompiled parameters.