Résumé
Download PDFnate.e.espejo@gmail.com www.nathanespejo.com github.com/HasNate618 linkedin.com/in/nathan-espejo
Education
Western University · London, ON
Bachelor of Engineering Science in Software Engineering, Dean's Honour List (2024, 2025) · Aug 2024 to May 2028 (expected)
Experience
Software Engineering Intern · TeraGo
May 2026 to Aug 2026Software Engineering InternMarkham, ON
- Co-led two data developers through Jira tickets and PR review on the print-approval dataset and benchmark.
- Built a labelled dataset from 4,620 scraped 3D models, rendered in Blender with labels checked by a VLM.
- Benchmarked small vision models as a binary safe-or-flag gate: 84.2% best agreement with VLM-checked labels.
VP of Data/ML (promoted from Software Developer) · 3D Western
Sep 2025 to Aug 2026VP of Data/ML (promoted from Software Developer)London, ON
- Built outage ticketing for non-standard provider emails: a local LLM (Ollama) extracts the key details and opens Jira issues through Microsoft Graph; works end to end, not yet deployed.
- Wrote Channel Finder, a Python tool that finds clear RF channels in MapInfo data, validated on 760 test cases.
- Deployed a Copilot Studio agent that answers engineers' RF questions from SharePoint documents.
- Proposed a self-hosted LLM platform (LibreChat with SharePoint and Jira) to cut per-user Copilot licensing costs.
Projects
Campus
Aug 2026 to nowsolo
- Python
- FastAPI
- SQLite
- embeddings
- cross-encoder reranking
- React
- Built a self-hosted AI study assistant that syncs my LMS courses into SQLite with Playwright and runs on any OpenAI-compatible endpoint, local or hosted.
- Designed hybrid retrieval over course files with embedding search, cross-encoder reranking and a lexical SQL prefilter. Fixed exact phrases that scored below the semantic cutoff.
- Wrote a 20-tool agent harness with an audit log for every AI write and a bounded tool loop, with 308 tests in CI.
- Use it across 6 courses for deadline lookups, studying from course slides, and reviewing drafts against briefs.
NixOS homelab
May 2026 to nowsolo
- NixOS
- Nix flakes
- Docker
- Tailscale
- Hermes Agent
- llama.cpp
- Defined a laptop and home server entirely in version-controlled NixOS config (27 hardened containers, Tailscale-only access, sops-nix secrets), so every change is a reviewable, reversible diff.
- Run one Hermes Agent profile per host that makes changes as config edits, over 472 sessions and 58,000 tool calls since June, with subagents taking larger investigations.
- Gated agent changes behind my approval and a nixos-safe-rebuild guard, with rollback to the last good build.
- Grew a git-versioned library of 191 agent skills on low-cost models, pruned weekly by a curator that keeps backups of removed skills.
AWARE
Jul 2026Hack the 6ix 2026solo
- Python 3.12
- YOLOv8n
- YAMNet
- Whisper-tiny
- MiniCPM5-1B (GGUF)
- Arduino UNO Q
- Built an on-device home-awareness system running YOLOv8 and YAMNet on a $80 Arduino UNO Q with QNN, with no cloud inference.
- Had an on-device LLM compile commands like "when glass breaks after 10pm alert me" into deterministic rules, then checked them every 500 ms with no LLM in the loop.
Hackathon Awards
8 awards across 22 hackathons.
- 3rd overall; Cohere's Challenge 1st place · NSBEHacks (Careerly)
- Best AI Application Built with Cloudflare · Hack Western 12 (Auralis)
- Best Use of MongoDB Atlas · TerraHacks (Lumen)
- Snap Spectacles AR Runner Up · Hack the North (Eyecandy)
- Overall 3rd place · Spark (CareBridge)
- Second Award · GameHACKS 1.0 (Street Cleaner)
- Best Useless Hack · CTRL+HACK+DEL 2.0 (Object-Love-Interface)
- Worst UX · Stupid Hackathon @ Western (Stupid Spotify)
Technical Skills
AI: LLM agents, tool calling, RAG, embeddings and reranking, MCP, VLMs, model evaluation, llama.cpp, Ollama, PyTorch, YOLOv8, ONNX
Languages: Python, TypeScript, C#, C++, SQL, Bash
Web & Infra: React, FastAPI, Next.js, NixOS, Linux, Docker, Tailscale, Caddy, Git, GitHub Actions
Tools: Hermes Agent, Microsoft Graph, Copilot Studio, Jira, SharePoint, Blender