Mitchell ArtzBuilding software &clarity from data.

I'm a full-stack developer and data analyst based in Ottawa, Canada. I build web and mobile applications end to end, engineer data pipelines and analysis, manage cloud infrastructure, and build geospatial and intelligence platforms, with cyber security applied throughout.

On the front end I use React, Next.js, TypeScript, and Astro; on the back end, Python (FastAPI, Flask, Django) and Node. I work with PostgreSQL/PostGIS and MySQL, Terraform, Docker, Kubernetes, AWS and Azure, Datadog, Flutter for mobile, and a broad GIS, data-science, and AI toolkit including GeoPandas, Pandas, scikit-learn, PyTorch, LangChain, and Ollama.

Yes. I live in Ottawa, Ontario, Canada, and work remotely with clients and teams across Canada and beyond. I currently develop remotely for Cinareo Solutions and have supported large studies for R.A. Malatest & Associates.

Web, mobile, and desktop application development; data analysis, ETL pipelines, dashboards, and business intelligence; AI and machine-learning applications (LLMs, RAG, agents, and computer vision); GIS, web mapping, and geospatial intelligence platforms; cloud infrastructure as code, DevOps, and observability; database design; cyber security consulting and hardening; and technical architecture advisory.

I have led transportation research and analysis on origin–destination travel datasets exceeding 1,000,000 trips, built ETL processes that ran up to 40× faster than their predecessors, applied statistical weighting against census profiles, and developed automation and reporting tooling in Python.

Yes, I take on select freelance and contract engagements. The best way to start is to email me a short description of the problem you are trying to solve, and I will point you in the right direction.

Counter-UAS (counter-unmanned aircraft systems) covers the detection, tracking, identification, and mitigation of small drones. Off-the-shelf UAVs are cheap, fast, and hard for legacy radar to see. I build software that fuses RF, acoustic, and optical sensors into a single air picture an operator can trust, turning noisy, disagreeing feeds into actionable tracks with geofenced alerting and threat scoring.

Open-source intelligence (OSINT) is insight drawn from publicly available data: social media, news, forums, public records, imagery, and tracking feeds. My platforms automate collection, resolve entities across messy sources into a single graph, surface hidden connections through link analysis, and present everything on a mapped, time-aware timeline with alerting for emerging situations.

My independent R&D focuses on domain awareness, Arctic sovereignty, travel-advisory monitoring, and counter-UAS tooling, areas adjacent to public-sector and national-security use cases. I am open to discussing scoped engagements where my skills in geospatial intelligence, data engineering, and secure infrastructure are a fit. Reach out and we can determine alignment.

I favour practical, grounded AI: retrieval-augmented generation over your own documents, tool-using agents for research and triage, and computer vision for classification tasks. I default to open-source, self-hosted models (Ollama, vLLM) when data sensitivity matters, and I treat AI as one component in a larger system, not a magic layer on top of bad data.

Yes. At R.A. Malatest & Associates I led an open-source business-process transformation, replacing proprietary time tracking, project management, and reporting tools with self-hosted alternatives and Apache Superset dashboards. I can assess your stack, map equivalents, plan the migration, and build the integrations so your team keeps working through the transition.

We start with discovery: mapping goals, constraints, data, and success metrics. Then architecture: a pragmatic plan with trade-offs written down. Build happens in tight, visible iterations behind version control and CI. Ship means reproducible deploys on infrastructure as code with observability and hardening. Support includes documentation, runbooks, and handover so your team can own it.

I typically reply within 24 hours on business days. For urgent production issues with existing clients, I prioritize same-day response. The fastest path is a short email describing the problem, any deadlines, and what a successful outcome looks like.

Yes. I build cross-platform mobile applications in Flutter: a single Dart codebase delivering native performance on iOS and Android. This includes offline-capable field data collection apps with local storage and sync, which I have shipped for large survey and research operations.

Transportation and travel-demand research, SaaS product development, market and social research, early-stage startup software, small-business IT and security consulting, and independent R&D across defence-adjacent technology (counter-UAS, OSINT, domain awareness). The through-line is data-heavy systems that need to be correct, fast, and maintainable.

Secure-by-design development, code review, infrastructure hardening, network segmentation, and observability. I run a home lab of honeypots and monitoring that captures real attacker behaviour and feeds back into hardening playbooks. Security is not a separate phase; it is woven into architecture, deployment, and operations.

Pricing depends on scope, timeline, and engagement shape: fixed project, monthly retainer, or advisory block. After a short discovery call or email exchange I provide a clear proposal with milestones and deliverables. No surprise invoices.

Yes. I routinely work with client research data, operational reports, and infrastructure credentials under NDA. I default to self-hosted and on-prem architectures when data cannot leave your network, and I document access controls and retention up front.

Often, yes. I onboard by reading code, running the stack locally, and shipping a small, visible fix or improvement in the first week. I match your git workflow, review culture, and deploy process rather than imposing a parallel one.

Yes, extensively. Large survey ETL, spatial joins, OD matrices, AIS track storage, and map tile backends. I design schemas and indexes for set-based geographic queries and pair Postgres with Python orchestration and Superset or custom front ends.

RF spectrum monitoring and direction finding, multi-sensor track fusion, YOLO-based visual classification, acoustic event detection, Remote ID ingestion, edge MQTT nodes, threat scoring, geofenced alerting, and realtime C2 dashboards, as modular software integrating COTS hardware.

Yes. I have deployed and administered Superset company-wide: connected sources, governed metrics, role-based access, custom dashboards, and training so teams adopt self-serve BI. I also help migrate from proprietary BI tools.

Yes. WebSocket-backed APIs, deck.gl and Leaflet front ends, PostGIS spatial queries, and time-aware replay for tracks and events. I design for sub-second updates where operators need them and graceful degradation when connectivity drops.

Seven years with R.A. Malatest & Associates including the Transportation Tomorrow Survey, one of North America's largest travel-demand studies. Million-record OD datasets, statistical weighting, PostGIS analysis, SurveyJS platforms, and Flutter field collection.

Yes. Architecture reviews, security assessments, performance profiling, Terraform and CI/CD audits, and data-pipeline correctness checks. Deliverables are prioritized findings with concrete remediation steps, not generic slide decks.

Yes, via monthly retainer or scoped support blocks. I prefer engagements that include documentation and handover so your team can own day-to-day ops, with me available for escalations, upgrades, and hardening.

I am based in Ottawa (Eastern Time) and work remotely with teams across Canada and internationally. I overlap with US Eastern and Pacific business hours routinely and align standups to your schedule for embedded engagements.

Source citation, retrieval evaluation, hybrid search, metadata filtering, human review queues for high-stakes outputs, and logging of queries and retrieved chunks. The LLM is one component in a verified pipeline, not the sole authority.

Yes. Webhooks, API integrations, CI/CD hooks, and notification routers are standard parts of platform work. I design integrations to be observable and idempotent so failures are visible and recoverable.

They are tailored to your sources, classification schema, and deployment constraints: self-hosted, with entity-resolution and link-analysis graphs you control. Commercial tools trade speed-to-demo for less flexibility; I build for your analysts' actual workflows.

I integrate COTS SDRs, cameras, and edge compute boxes and write the software stack around them: detection, fusion, MQTT transport, and dashboards. I do not design custom PCBs, but I deploy and harden field hardware running Linux and Docker.