Junior Software Engineer @ Green Tanner Industrial  |  MS Data Science @ Arizona State University  |  Full-Stack & AI Builder

I finished my MS in Data Science, Analytics and Engineering at Arizona State University in May 2026 and now work as a Junior Software Engineer at Green Tanner Industrial. I've built production software systems, ML pipelines, and full-stack applications across internships, a full-time role, and side projects.

My work sits at the intersection of software engineering, applied ML, and data engineering. I care about building things that actually work in production, not just in notebooks.

• Junior Software Engineer at Green Tanner Industrial (Chandler, AZ), building full-stack features and a RAG pipeline for an internal construction-management platform
• Building Harper, a terminal coding agent in Go that works with five LLM providers
• Built a self-healing SRE agent from scratch that fixes CI/CD failures on its own, with a 95% fix success rate
• Regular open-source contributor to GRASS GIS (OSGeo)
• Based in Tempe, AZ

• Build and maintain full-stack features for an internal construction-management platform, connecting field and office workflows through API integrations with tools including Procore
• Develop a Retrieval-Augmented Generation (RAG) pipeline for internal software, using retrieval and semantic search in a production AI system
• Work across Rust, WebAssembly, HTMX, Alpine.js, Tailwind CSS, and PostgreSQL, in a pair-programming and mentorship model with senior engineers

• Researching a community-driven tourism and entertainment discovery platform for Central Alberta, focused on local events, attractions, transportation access, and grassroots economic development
• Supporting early-stage research on data sources, geospatial analysis, and AI-assisted workflows to improve tourism visibility for smaller communities

• Architected Python backend microservices and REST APIs for an AI journaling app's transcription and emotion-analysis pipeline, reducing inference latency by 21% across 30+ concurrent data streams during 200% month-over-month growth
• Shipped a React and Flask dashboard for monitoring Whisper transcription and journaling insight workflows, improving production visibility and cutting incident response time by 35%
• Built an offline processing pipeline that eliminated data loss in journaling entries during connectivity failures, and implemented OAuth2 authentication that improved token verification speed by 27%

• Built scalable Python, Django, and PostgreSQL backend services for a documentation platform, tuning SQL indexing and query performance to speed up responses by 18% across core features
• Streamlined CI/CD pipelines with Docker, Kubernetes, and GitHub Actions, cutting deployment errors by 30% across 5+ services
• Worked with QA on debugging and unit testing to trace and fix REST API integration failures, reducing recurring defects by 25%

• Automated email response workflows and redesigned the company website in PHP with SEO optimization on AWS, reducing manual response time by 30% and increasing organic search traffic by 20%

Regular contributor to GRASS GIS (OSGeo/grass), a free and open-source geospatial processing engine used for raster and vector analysis, terrain and image processing, and spatial modeling. It is written mostly in C and used by researchers worldwide. I mainly contribute test coverage for its modules and C libraries.

Go · Ollama · vLLM · Anthropic

A provider-agnostic terminal coding agent written in Go. It supports five LLM providers, including Anthropic, and runs an orchestrator/worker loop that hands subtasks to workers in parallel through concurrent goroutine fan-out.

• Permission system with three modes (allow, ask, deny) and structured JSONL session logging, so the engineer stays in charge of what the agent is allowed to run

LangGraph · Groq Llama 3.3-70B · FastAPI · Streamlit · PyGithub · LangSmith · Python AST

Multi-agent AI system for automated incident response. Built from scratch on nights and weekends because the problem was worth solving. V1 is complete and the core self-healing loop works end to end.

• Multi-agent orchestration via LangGraph, with Investigator and Mechanic agents that hand off explicitly and track their communication
• Self-correcting loop that retries up to 3 times with validation feedback, the way an engineer would debug
• 95% success rate after self-correction, resolving incidents in 30-60 seconds at $0.02-0.06 per fix
• Full observability through LangSmith decision tracing and GitHub Actions logs
• Safe by default with iteration limits, AST-based code validation, and human approval required before merge

Next.js · FastAPI · Neo4j · Mistral-7B · LoRA · Ollama · Docker · GCP

Full-stack mental health chatbot with a Next.js frontend and a FastAPI backend exposing JWT-authenticated REST APIs, backed by a Neo4j graph database. Containerized with Docker and deployed on GCP.

• Fine-tuned Mistral-7B-v0.3 with LoRA and supervised fine-tuning, served through Ollama
• Benchmarked against a curated evaluation set to reach 91% accuracy at sub-500ms latency

Python · FastAPI · TensorFlow (Keras) · Scikit-Learn · Machine Learning

FastAPI backend with a dynamic multi-model ensemble for Remaining Useful Life (RUL) prediction. It ingests and aggregates predictions across Keras (.h5) and Scikit-Learn (.pkl) models for fleet health monitoring, and a React dashboard shows real-time risk scores and predictive alerts.

Python · TensorFlow/Keras · U-Net · Google Earth Engine · Sentinel-2

Deep learning pipeline for pixel-wise land use classification over Central Phoenix from Sentinel-2 satellite imagery. A custom U-Net covers 9 land cover classes, with patch-based training and full-image reconstruction.

• 95.95% pixel accuracy, 95.50% weighted F1, 92.79% mean IoU

Python · Data Analysis · Cosine Similarity · Recommender Systems

User-based collaborative filtering recommender trained on the Book Crossing dataset. It computes cosine similarity across user rating vectors to find similar readers and suggest books.

• M.S. Data Science, Analytics and Engineering, Arizona State University (Aug 2024 – May 2026)
• B.Tech, Computer Science and Engineering, Nirma University (Oct 2020 – Aug 2024)

Feel free to reach out on LinkedIn or by email.