[ esc / enter / space to skip ]

joshua@portfolio:~ — login
joshua@portfolio:~$whoami

Joshua Gomes

Sr Software Engineer @ Qualtrics · Vancouver, BC

Building LLM observability platforms, configurable RAG infrastructure, and HIPAA-compliant voice agents at scale. Off-hours I'm hiking, on a guitar, or chasing a soccer ball.

$ find . -socials
joshua@portfolio:~/about $ htop
joshua@portfolio · up 3y, 2w · load avg: 0.87, 0.60, 0.42 · tasks: 1 running, ∞ queued
profile

./profile.jpg

# bio

I'm Joshua — a senior software engineer specializing in AI platform engineering. At Qualtrics I'm on the AI enablement team: I've built a self-hosted LLM observability platform on GCP, a configurable RAG service that consolidated retrieval infrastructure across 15+ teams, a HIPAA-compliant voice agent for proactive customer outreach, and a full-stack survey automation platform that cut processing time by 96%. Before that, AI-driven virtual care at Codex Health and big-data privacy infrastructure at Affinity Solutions. I'm at my best when AI features have to be both ambitious and reliable enough to put in front of real users.

// metadata

  • education:University of California, Irvine
  • interests:Hiking, Soccer, Photography
  • location:Vancouver, BC
  • employment:Qualtrics

// now playing

  • caffeine78%always brewing
  • curiosity100%unbounded
  • free time22%see you on a trail
  • guitar47%strictly bedroom
joshua@portfolio:~/resume $ tmux

// experience.log

Senior Software Engineer

@Qualtrics
Jan 2025 - Present·Vancouver, BC

Qualtrics is the experience management platform used by thousands of enterprises. I'm on the AI enablement team — building the shared infrastructure that lets every product team ship LLM features faster and with more confidence.

  • Identified the lack of centralized LLM observability and proposed self-hosting Langfuse on GCP (Cloud Run, Redis, ClickHouse), giving 5+ product teams unified visibility into traces, evals, and quality metrics.
  • Built a configurable RAG service on TurboPuffer for Qualtrics' internal LLM platform, exposing chunking strategies, embedding model selection, metadata enrichment, hybrid search, and reranking — consolidating retrieval infrastructure across 15+ teams that had defected to Azure AI Search.
  • Engineered a full-stack survey automation platform (C#, Azure) for HCAHPS compliance, integrating an AI config layer that reduced per-request processing time from 2 hours to 5 minutes — 96% reduction across ~50 specialists handling hundreds of daily requests.
  • Built a HIPAA-compliant proactive voice agent (Twilio, Pipecat, Claude) on AWS to contact customers after low satisfaction scores — reduced time-to-resolution from 3 days to under 30 minutes, 65% first-call resolution rate.
  • Spearheaded GitHub Copilot adoption across the engineering org (PR reviews, code generation loops, prompt workflows) and delivered an internal skill marketplace — cut PR review time by 35% and boilerplate dev time by 50%.

Software Engineer

@Codex Health, Inc.
September 2022 - December 2024·Palo Alto, CA

Codex Health builds AI-driven virtual care — tools that help clinicians deliver better outcomes at lower cost.

  • Re-architected the GCP serverless stack for multitenancy — meaningful cost reductions and headroom to onboard more clinics without per-tenant overhead.
  • Led implementation of population-health analytics, surfacing the signals clinicians needed to drive proactive, personalized interventions.

Software Engineer, Big Data

@Affinity Solutions, Inc.
August 2019 - August 2022·San Jose, CA

Affinity Solutions sells consumer purchase insights at scale. I worked on data privacy infrastructure and the big-data pipelines feeding the platform.

  • Built the DSAR / CCPA compliance framework (Python, FastAPI) — RESTful APIs for data subject access requests at consumer scale.
  • Migrated the DSAR ETL pipeline to AWS, improving performance and retiring a fragile on-prem dependency.
  • Optimized AWS EMR clusters and Airflow-orchestrated pipelines on EC2 — 20% reduction in ETL processing time.

// education.log

Bachelors of Science, Computer Science

Computer Science·University of California, Irvine (UCI)

Focus: Information Systems
Relevant Coursework: Machine Learning, Neural Networks and Deep Learning, Artificial Intelligence, Search and Information Retrieval, Human Computer Interaction, Principles in System Design, Design and Analysis of Algorithms, Algorithms and Complexity, Advanced Network Security and Cryptography, Computer Systems and Assembly, Database Management Systems

joshua@portfolio:~/projects $ ls

// projects

4 entries
LeadSwitchboard[ preview ]

./leadswitchboard

Operations platform for pay-per-lead agencies — automates buyer routing, real-time follow-up, and the operational glue between lead gen and closing. Replaces the spreadsheets-and-Slack-DMs status quo with a real-time pipeline so agencies stop bleeding revenue to slow follow-up and manual buyer ops.

Lead RoutingBuyer OpsAutomationSaaS
AskMyDocs[ preview ]

./askmydocs

GenAI-powered document chatbot (RAG): upload PDF/DOCX/CSV, chunk and embed with OpenAI, store in Supabase pgvector, and chat grounded answers via a Next.js 15 backend.

Next.jsTypeScriptSupabasepgvector
Nesso[ preview ]

./nesso

Nesso is an iOS app designed to help you maintain meaningful connections with your contacts by managing and scheduling regular reminders to reach out.

Swift
PrimerJobs[ preview ]

./primerjobs

Two-sided marketplace for house painting. Consumer surface (primerjobs.com) books homeowners with vetted painters at $0 deposit; contractor HQ (hq.primerjobs.com) lets painters manage jobs, customers, and their pipeline. Both surfaces built end-to-end.

Two-Sided MarketplaceBookingDashboardSaaS
joshua@portfolio:~ $ ./connect.sh

Get in touch.

I'd love to connect! Whether it is to discuss a job opportunity, a personal project, or just life, please reach out!