This artifact comes from AIML‑500 Workshop 1, where the assignment was to research the historical development of artificial intelligence and machine learning and present it as a timeline. A timeline invites a book report: a list of dates and names in order. I decided the presentation would be stronger as an argument, so I built it around a single thesis: every major leap in AI, from the artificial neuron in 1943 to today's large language models, was unlocked by compute, data, and distributed systems as much as by algorithms. The ideas consistently arrived decades early; what changed at each breakthrough was the substrate underneath them.
The result is a seven‑slide presentation. It opens by naming the three forces that gate every breakthrough on the timeline: compute, data, and distributed systems. It then walks three eras: the foundations era (1943 to 1973), where the core ideas arrive but stall into the first AI winter for lack of hardware and data; the data‑driven turn (1980 to 2011), where expert systems collapse, backpropagation waits on hardware, and the web finally makes data abundant; and the deep learning and generative era (2012 to 2026), where the three forces align and progress goes vertical. Each era closes with a synthesis of why progress stalled or accelerated, including a parallel from my own field: fraud detection became viable not because of a cleverer algorithm, but because transaction data went digital and streaming infrastructure could score it in real time. The final content slide turns the thesis forward, to compute concentration, data quality as the new ceiling, and responsible stewardship, and the deck closes with a full APA 7 reference slide.
The argument in one view:
The assignment objective was to demonstrate an understanding of how artificial intelligence and machine learning developed over time and to communicate that history clearly. My personal objective was to practice a different skill than Artifact 01 demonstrates. That artifact shows I can design and ship an AI product; this one shows I can synthesize a technical domain into a defensible thesis and communicate it to a mixed audience. I set two rules for myself: every era had to advance the argument rather than list events, and every claim had to trace back to the sources on the reference slide.
The presentation went through five stages:
The biggest challenge was turning more than eighty years of events into a seven-slide argument without reducing the work to a list of dates. I learned to choose milestones by explanatory value: each slide had to show how an algorithmic idea became practical only when compute, data, and infrastructure were ready. The resulting thesis-first structure is a method I now use for technical briefings.
This is not a chronology of AI; it is an argument about why AI advances, and the third force in that argument, distributed systems, is my day job. It demonstrates the research synthesis and technical communication skill that complements the product‑building skill in Artifact 01, and it is the same skill behind my public writing on database internals.
Organizations deciding where to invest in AI need engineers who can explain why capability jumps happen and what infrastructure each one requires, in language executives and engineers both accept. Senior backend and forward‑deployed roles are half communication; this artifact is evidence I can do that half.