Getting our self-sufficiency up and healthy

Self-sufficient systems (and people) are the foundation for us (and any scaleable organisation). Needless to say self-sufficiency is existential for us not just to survive but thrive. Mid-2024 (early AI-wave), we started investing a large share of our time into building the muscle that gets us AI-ready if not AI-native, without hurting our BAU growth.

The key problems can be classfied into -

• A pile of tribal context and tooling
• Numerous repositories and services, aka debt
• Scaling challenges - be it our queues, api gateways or data pipelines
• Leaky knowledge in monitoring, debugging and handling/resolving on-calls/escalations
• Staying compliant - GDPR, SOC-2 and not using any random model/API out-there
• Lastly - the mystical nature of Clojure (which is odd to me but it’s also clear that very few seem to enjoy unless you already like Clojure)

How we went about solving those iteratively and layer-by-layer - From a 3-person on-call roster to a 0-person defense layer at a larger scale and a siginficant set of side benefits enabling the whole org across growth, sales and other product functions (ie our AI-first ops engine)

What you should expect

• Working backwards towards self-sufficiency state, AI-native
• Where we struggled and how you can avoid it, ie jump to running, avoid the crawl (IRL struggles at scale)
• Becoming a git commit org, ie everyone at Swym is working on/via GitHub

Who might find it interesting

You have a product/platform built pre-AI and want to scale your product/teams to be as AI-native as possible without burning yourself or your budget/tokens

Who am I

An early engineering employee at Swym Corp for the past 10-ish years. Swym helps ~50K merchants(doing ~$50B+/yr in GMV) across the globe capture intent and convert effectively in the ecommerce/omnichannel space. I/we build whatever that problem/opportunities need across the stack. Before that, early employee at CloudPact (acquired as Mowbly by ASG/Rocket Software) building out a RAPD platform enabling enterprises to build/deploy/manage their mobile initiatives pre-2015.

If you need a title, it’s VP of Engineering, but we all know that is irrelevant.

Draft deck

Roughly generated using AI with more internal context - https://docs.google.com/presentation/d/1cgQ7TwELbrns4Qw1AknMtAaq-S1gXPqNLZZcgZo5wHs/edit?usp=sharing

Misc context

More for the speaking style - https://hasgeek.com/inclojure/2020/sub/clojure-macros-the-good-the-bad-and-the-messy-CHMjgNqrUVknxHRcHW1JtK - I am slightly older now though :)
The blogs/podcasts close to the narrative I intend to cover in more detail

1. https://stories.getswym.com/p/recap-the-great-library-merge-a-journal
2. https://stories.getswym.com/p/how-swym-prepares-for-the-biggest
3. https://stories.getswym.com/p/inside-bfcm-real-time-ops-monitoring
4. https://stories.getswym.com/p/from-zero-to-production-lessons-from