Anshudeep Mathur

GitHub LinkedIn Email

anshuweb.comsprine @ github[email protected]+1 (226) 600-8755

Technical skills over time ★ = actively practiced.
Bolded = I think I'm good at this. 

                                                                                 ★ Swift / SwiftUI
                                                                              ★ Cloudflare Workers
                                                                           ★ MCP servers
                                                                        ★ LLM evaluation and grounding
                                                                     fine-tuning, DSPy, local embeddings
                                                                  ★ LLM APIs and open-weight models
                                                               ★ agentic engineering method
                                                            Blender
                                                         ★ three.js, GLSL
                                                      ★ TypeScript
                                                   ★ Prototyping
                                                GCP infrastructure
                                             ★ DuckDB, SQLite
                                          ★ d3.js
                                       ★ reshaping and transforming data
                                    data analysis in Python (pandas, statsmodels)
                                 event streaming using Spark
                              scala
                           react
                        ruby on rails
                     spark
                  elasticsearch
               sas text miner
            oracle text
         oracle pl/sql and etl
    ★ python
  ★ sql
★ web dev + css


------------▴--------------▴------------------▴------------------▴-------▴-------▴-------▴-------▴ today
            2011           2015               2020               2023    2024    2025    2026

Technical skills

AI and LLM. LLM evaluation and grounding, LLM APIs and open-weight models, agentic engineering method, MCP servers, fine-tuning, DSPy, local embeddings.

Data. sql, oracle pl/sql and etl, oracle text, python (pandas, statsmodels), DuckDB, SQLite, reshaping and transforming data, spark (PySpark, event streaming), elasticsearch, sas text miner.

Build. Prototyping, d3.js, web dev + css, TypeScript, three.js, GLSL, Swift / SwiftUI, Cloudflare Workers, react, ruby on rails, scala, Blender, GCP.

Work experience

Founder, Galt Software galt.software · Cambridge, Ontario April 2026 — present
Staff Software Engineer, The Catalog Shopify March 2025 — March 2026
  • Shopify's venture into agentic commerce: an opinionated product knowledge graph turning fragmented, free-form merchant data into standardised entities and attributes.
  • Designed and productionised LLM inference pipelines at the scale of millions of inferences: fine-tuned Qwen 3, wrote LLM evaluations and judges, built continuous evaluation for data quality and drift, set up MCP tooling.
  • Collaborated across product, infrastructure and data teams on schema, embeddings and metadata alignment.
Staff Software Engineer, People Analytics Shopify 2022 — March 2025
  • Responsible for the systems that touch people data across Shopify, and for how teams collaborate on sensitive data without exposing it; mentored engineers and analysts.
  • Advanced analysis pipelines for free-form text responses using statistical method and LLMs.
  • Prototyped design, planning and visualisations for understanding Shopify's employee population and re-organisation, and for impact and performance strategies.
  • Treated compensation fairness, fairer performance expectations and headcount modelling as engineering and systems problems for executive decisions.
Senior Data Engineer, People Analytics Shopify April 2019 — 2022
  • Led the evolution of Shopify's Talent data platform development: consolidating extraction, modelling and observability over sensitive employee data (hiring, performance, compensation, departures), and with Shopify's wider data ecosystem, including the move to BigQuery.
  • Built survey tooling for the annual engagement survey, a trickle survey for a steadier signal, hierarchy-scoped reporting so a lead sees only their own team, and a correlation explorer relating survey questions to outcomes.
  • Mentored engineers and analysts through the move to remote work.
Earlier roles at Shopify
  • Senior Data Engineer, Data AppsAugust 2017 — April 2019The Data Portal for discovering analyses (see shopify.engineering, "Solving data discovery challenges at Shopify"); a Kafka to Amplitude stream at thousands of events a second.
  • Data Engineer, Shopify PlusSeptember 2015 — August 2017Spark and Kafka data models in production; introduced Jupyter notebooks to the data organisation.
Data Analyst and Architect, Quality BlackBerry, Waterloo 2008 — 2010 (co-op), January 2011 — September 2015
  • Built a full-text search engine over two million internal documents with Oracle Text, and later tuned retrieval over about 2 GB for thousands of daily users.
  • Voice of the Customer: NPS and satisfaction analysis for executives; an Elasticsearch, Python and React tool to query verbatims by country, product and question; text mining with SAS Text Miner, then Python.
  • Ran experiments to evaluate designs for a data-mining platform (interfaces through to infrastructure) that shaped a cross-functional Six Sigma project.
  • Taught Python, Elasticsearch and Git to the analytics team so SAS and SQL processes could move to Python.
Earlier roles at BlackBerry
  • Data Analyst and Architect, Quality2015
  • Analyst, Voice of the Customer2014
  • Quality Manager, Enterprise2014
  • Database Applications Analyst2012 — 2014
  • Database Systems Developer2011 — 2012
  • Junior Quality Database Developer and Analyst (co-op)2008 — 2010
Web developer (part-time), IT Project Management Toronto District School Board May 2007 — May 2008
  • Redesigned and launched the public home page (about 160,000 visitors a month); accessible ASP and .NET development; Google Maps school profiles.

Education

BAComp, Applied Computing Wireless technology and computer science, co-op. University of Guelph-Humber 2006 — 2010

Before and during university

Other