Humza Shah
Introduction
I am a hands-on engineering leader, with demonstrable expertise in creating scalable software and infrastructure for high-traffic public-facing always-on projects.
Online presence
- humzashah at gmail dot com
- https://github.com/humzashah
- https://stackoverflow.com/users/2493089/shs
- https://www.linkedin.com/in/shumzash/
Work Experience
Staff Software Engineer, Manager at Credit Karma, London
Since July 2019
I joined as Credit Karma UK’s first staff engineer and was attached to multiple teams to provide software and platform infrastructure leadership.
In my first year, I led the migration of workloads and customer data after the acquisition of Noddle.
By mid-2021, I had completed the UK team’s service mesh migration from Linkerd to Envoy, led the implementation of GDPR-compliance software, and led data pipelines for a number of internal systems and external (Intuit). A lot of my time was also spent architecting platform security applications that served PII and that encrypted data.
Since July 2021, I have been the line manager of the UK Platform team. I lead six engineers covering compute, storage, logging, observability, and developer experience for Credit Karma UK.
Senior Software Engineer at Deliveroo, London
From September 2016 to July 2019
At Deliveroo, I played an instrumental role in multiple projects.
In the Logistics Algorithms team, I worked closely with data scientists on Deliveroo’s famous rider dispatch algorithm - Frank; soon afterwards, I led the design and development of the company’s first framework to train statistical models - Canoe.
In the platform team, I reduced datastore costs, built load-source observability, and helped transition from monolith to service oriented architecture.
My last few months were spent in the Payments group, adding support for new payment providers and improving the build-times of the central monolith.
Tech lead at Innovation Enterprise, London
From September 2015 to August 2016
I secured IE’s online content, set up reliable continuous deployments for online services, and built the company’s first data pipeline consolidating its various data sources.
Software Engineer at OmniScriptum, Saarbrücken
From May 2014 to September 2015
I worked in a small team to develop features for multiple Ruby-on-Rails websites, used by authors to submit books to be published.
Software Engineer at ProtonLabs, Lahore
From November 2012 to April 2014
I joined senior engineers to build and maintain HedgePo (now Allocator). The product connected hedge fund managers with investors; we built it using Ruby-on-Rails, MySQL, Heroku, and CoffeeScript.
Conference Talks
Improving the reliability of a monolithic web service
At USENIX SREcon, Singapore in June 2019
Automating development workflows using GitHub features
At GitHub Constellation, London in March 2018
Training statistical models in containers with Kubernetes
At Kubernetes Meetup, London in February 2018
Education
B.Sc. Electrical Engineering
University of Management and Technology, Lahore
Fall 2008 - Spring 2012
GPA: 3.74 on a scale of 4.00
Rector’s Merit Award, Spring 2009
Dean’s Merit Award, Fall 2009
President Science Society, 2011
Professional courses
Operations Training for Kafka
By Confluent, London in April 2018
Python Data Science Bootcamp
By Cambridge Coding Academy, London in July 2016
Advanced Linux Administration
By City University, London in April 2016
Competencies
I have at-scale production experience with the technologies listed below; entries with asterisks are amongst my favourites.
Amazon Web Services*, Angular, Ansible*, Chef, CircleCI*, Cloudflare*, CoffeeScript, Datadog*, Docker*, DynamoDB, Finagle, Flask, Git*, Go, Google Cloud Platform, Grafana*, GraphQL, HAProxy, Heroku, HSM, JavaScript, Jenkins, Kafka, Kinesis, Kubernetes, Memcached, MySQL, New Relic, Nginx*, PgBouncer, PostGIS, PostgreSQL*, Protobuf, Python, Rails*, React, Redis*, Redshift, REST*, Ruby*, Salt, Scala, Segment, Sentry*, Snowflake, Splunk, Terraform*, Thrift, Travis