Andre Graham
Toronto, ON M8X 1L5 · adregrhm@gmail.com
Objective
AI Engineer and Data Science graduate who builds secure, multi-agent LLM systems to automate decision-making and unlock data-driven value. I specialize in deploying complex AI models—from machine learning to agentic workflows—within robust, end-to-end web applications.
Key Qualifications
- Delivered end-to-end applications using the Flask factory pattern, with back-ends powered by PostgreSQL/SQLAlchemy and key integrations for Adobe Sign & Google APIs.
- Architected planner→solver→critic multi-agent workflows using Autogen & GPT‑4o.
- Designed GDPR/PIPEDA-compliant systems featuring AES‑128 encryption & RBAC.
- Hands-on ML modelling with scikit-learn, XGBoost, TensorFlow, Keras and more.
Technical Skills
Languages
Python · SQL · JavaScript · HTML
Full Stack Development
Flask · SQLAlchemy · PostgreSQL
Data & ML
Pandas · Scikit‑Learn · XGBoost · TensorFlow
DevOps / Cloud
Docker · Azure · Redis · Celery · Local Servers
Education
Graduate Certificate – Applied AI Solutions Development
George Brown College · GPA 3.95/4.00 · Academic Excellence Award (top of class)
BSc – Statistics & Economics (Distinction)
University of Toronto · GPA 3.39/4.00
Professional Experience
AI Engineer (Contract) · Kinectrics
- Architected multi‑agent AI platform (Flask, Autogen, GPT‑4o) enabling increased producivity in reporting.
- Worked on creating minimum viable products used to pitch new project ideas.
- Dynamically adapted to constantly changing products.
Automation Consultant (Contract) · Summerhill Capital Management
- Lead developer of ClientVault: Flask‑PostgreSQL platform automating KYC onboarding & compliance reminders.
- Integrated AES‑128 encryption, BitLocker‑backed Docker volumes & VPN‑gated access (GDPR/PIPEDA).
- Built Celery scheduler emailing Excel & HTML digests, removing the necessity for all manual checks.
- Introduced diff‑logging utility for full audit trails.
AI Engineer – Student · Kinectrics
- Prototyped multi-agent models leveraging Autogen and OpenAI's API to develop MVP used to sell AI based enhacment to productivity.
Senior Instructor · University of Toronto
Camp Instructor · University of Toronto
Painting & Staining · Student Works
Highlighted Projects
ClientVault
Secure client‑data vault with RBAC, AES‑128 encryption & Celery reminders.
Multi-Agent Systems
Planner‑solver‑critic LLM agents orchestrated via Autogen; 3× faster reports.
Interests
Soccer · Basketball · Guitar · AI Research
Let's Collaborate