JACOB MACINNIS

Tech Leader | Mentor

Cloud & AI Enthusiast


About Me

Jacob MacInnis
  •   Hands-on engineering leader with deep experience in architecture, development, and technical leadership. Proven track record of designing and deploying scalable, cloud-native applications. Adept at leading engineering teams, fostering innovation, and optimizing processes.
  • Completed AI/ML certifications from Stanford, DeepLearning.AI, Duke, and Google, combining strong theoretical foundations with practical implementation. Built and deployed models for image recognition, classification, NLP, transfer learning, RAG, and AI agents. Experienced with TensorFlow, PyTorch, scikit-learn, LangChain, and LangGraph.
  • Published 50+ technical articles on Medium covering TypeScript, Python, data structures, algorithms, and applied machine learning.


  • What I Do Best:
  • Architect Scalable Cloud Solutions – Specializing in TypeScript, Node.js, React, AWS serverless
  • Build & Mentor Engineering Teams – Guiding developers at all levels to achieve technical excellence
  • Align Tech with Business Goals – Ensuring engineering efforts drive measurable business impact
  • Deliver Practical AI Agents – Designing LangGraph-powered tools with LLM reasoning, memory, and tool use
  • Modernize Legacy Systems – Transforming monoliths into resilient, cloud-native architectures


  • Tech Stack & Interests:
  • 💡 Experienced in: TypeScript, JavaScript, Node.js, Python, React, Redux, FastAPI, LangChain, LangGraph, Ollama, OpenAI, AWS, GCP, Azure, Firestore, SQL, NoSQL, Docker, Poetry
  • 💡 Passionate about: AI/ML Agents, ML Pipelines, Full-Stack + ML Systems, Engineering Leadership

  • Leadership Philosophy:
  • 🔹 Technical Excellence – Staying hands-on to ensure architectural integrity and best practices
  • 🔹 People Development – Mentoring and fostering growth at all levels
  • 🔹 Strategic Vision – Driving decisions that align technology with business outcomes


  •   Recent projects include autonomous agents, LLM-based refactoring systems, multi-model orchestration, and Retrieval-Augmented Generation (RAG) APIs. I work hands-on with tools like LangGraph, FastAPI, SentenceTransformers, T5, FAISS, and Docker to ship fast, useful prototypes and production tools.


  •   I hold certifications from Stanford, DeepLearning.AI, Duke and Google in Machine Learning, AI Agents, Managing Machine Learning In Production AI Business Strategy, and Data Analysis, blending emerging technologies with real-world software solutions. I thrive in just-in-time architecture to enhance product agility and delivery.


  •   I regularly share insights on TypeScript, Python, AI/ML, and modern software architecture to educate and inspire the tech community. Looking to connect with fellow technology leaders who are passionate about engineering leadership, cloud architecture, and AI-driven solutions.

Machine Learning Projects (Made in 2025)

R2AI Agent

R2AI is a semi-autonomous refactoring agent built using LangGraph. It analyzes and improves Python code structure using Ruff, refactors it via LLM, and supports human-in-the-loop review with feedback that can trigger iterative improvements.

R2AI Agent screenshots

Technologies Used

Python, LangGraph, LangChain, Ollama, Mistral, OpenAI API, Ruff, Poetry

Agent Cortex v2

Agent Cortex v2 expands the local AI assistant with new tools including: persistent long-term memory, Python code execution, and short-term conversational context. Built on LangChain and Mistral 7B, it enhances the original RAG, math, and web search tools

Agent Cortex v2 screenshots

Technologies Used

Python, LangChain, Mistral, Ollama, ChromaDB, FAISS, DuckDuckGo Search, Python Code Execution, Poetry

Agent Cortex v1

Agent Cortex is a local AI assistant powered by LangChain and Mistral 7B. It combines: Document retrieval (RAG), Math solving, and DuckDuckGo web search. Runs fully offline (except for websearch)—no paid LLM APIs

Agent Cortex v1 screenshots

Technologies Used

Python, LangChain, Mistral, Ollama, ChromaDB, FAISS, DuckDuckGo Search, Python Code Execution, Poetry

RAG API

production-ready RAG (Retrieval-Augmented Generation) backend built with FastAPI, SentenceTransformers, and FLAN-T5. It demonstrates a modern approach to question-answering using both semantic search and language generation.

RAG API screenshots

Technologies Used

Python, FastAPI, SentenceTransformers, transformers, torch, numpy, FAISS, FLAN-T5, Docker, Poetry

Agricultural Leaf Vision

A lightweight computer vision project developed to deepen my understanding of image preprocessing, transfer learning, and end-to-end ML deployment with minimal cloud cost. It's built using transfer learning with CNN EfficientNetB2

Agricultural Leaf Vision screenshots

Technologies Used

Python, TensorFlow, Keras, scikit-learn, EfficientNetB2, Transfer Learning, matplotlib, seaborn, firebase, google-cloud-aiplatform, Docker, FastAPI, Poetry

Summit Mind

a portfolio project built to experiment with and deploy lightweight abstracted summarization models based on the T5 (Text-to-Text Transfer Transformer series) models. Designed to summarize multi-turn conversations (e.g., meetings, support chats)

Summit Mind screenshots

Technologies Used

Python, PyTorch, T5, Hugging Face Transformers, Numpy, FastAPI, Firebase, Google Cloud Run, Google Cloud Registry, Docker, Poetry

Career Compass

A production-grade machine learning system that predicts whether a user is currently employed based on structured profile attributes. It uses a supervised binary classification model built with TensorFlow/Keras (Sequential API) — a deep feedforward neural network trained on structured tabular data.

Career Compass screenshots

Technologies Used

Python, TensorFlow, Keras, Pandas, Scikit-learn, Matplotlib, Seaborn, FastAPI, PostgreSQL, Cloud SQL, Google Cloud Run, Google Cloud Registry, Docker, Poetry

Tech Stack

Projects (Made in 2019)

Function Earth

Available at Google Play Store - React Native application t​racks what we are doing around the world to reduce our footprint and preserve our environment for future generations. Built a compost bin? Recycled? Picked up trash from the ocean? Helped the elderly? Took care of an animal? Use Function Earth to log these actions. Race other cities, states and countries to be the greenest region in the world!

Function Earth screenshots

Technologies Used

JavaScript, Node, React-Native, React, Redux, Express, MongoDB, Mongoose, mLab, HTML5, CSS3, Passport-JWT, Jest, Mocha, Chai, Enzyme, Travis-CI, Heroku

Beach Tides Mobile App

Available at Google Play Store - MVP React-Native Application Beach Tides is built to give real time and future tide predictions for the coastal United States. Use BeachTides to search by zip code or city and get beach tide predictions. Search for tide information without creating accounts

Beach Tides Mobile App screenshots

Technologies Used

TypeScript, JavaScript, Node, React, Redux, Express, Socket.io, MongoDB, Mongoose, mLab, HTML5, CSS3, Passport-JWT, Mocha, Chai, Enzyme, Travis-CI, Heroku

Think Programming

A place to challenge other coding enthusiests in languages or programing categories. Be fastest and cleanest coder to level up fast. Gain experience and coding muscle memory fast.

Think Programming screenshots

Technologies Used

Node, React, Redux, JavaScript, Express, Socket.io, MongoDB, Mongoose, mLab, HTML5, CSS3, Google-OAuth-2.0, Passport-JWT, Mocha, Chai, Enzyme, Travis-CI, Heroku

Beach Tides Web App

A Full-Stack mobile first responisive web app built to give users in coastal U.S. tidal predictions. Search by zipcode or city and state. Create accounts using Google OAuth 2.0 to save locations and more.

Beach Tides Web App screenshots

Technologies Used

Node, React, Redux, JavaScript, Express, MongoDB, Mongoose, mLab, HTML5, CSS3, Google-OAuth-2.0, Passport-JWT, Mocha, Chai, Enzyme, Travis-CI, Heroku

CS Cards

Using a spaced repetition algorithm CS Cards teaches users Computer Science Concepts. It is built using linked-lists in a mongo db and changes values of data structures depending on user answers.

CS Cards screenshots

Technologies Used

Node, React, Redux, Socket.io, WebSockets, JavaScript, Express, MongoDB, Mongoose, mLab, HTML5, CSS3, JWT-Auth, Passport-JWT, Mocha, Chai, Enzyme, Heroku

Scottish Gaelic

Scottish Gaelic is a language learning app designed to improve your abilty to learn the beautiful Scottish language of Gaelic. Using our specialized algorithm it teaches words and phrases.

Scottish Gaelic screenshots

Technologies Used

Node, React, Redux, Socket.io, WebSockets, JavaScript, Express, MongoDB, Mongoose, mLab, HTML5, CSS3, JWT-Auth, Passport-JWT, Mocha, Chai, Enzyme, Heroku

Prime Three

One day project built to find the first prime number that contains the three digits a user inputs.

Prime Three screenshots

Technologies Used

Node, React, Redux, JavaScript, Express, MongoDB, Mongoose, mLab, HTML5, CSS3, Mocha, Chai, Enzyme, Sinon, Heroku