Simor

Data Infrastructure for Production AI

Practical writing on AI data engineering, feature stores, and the infrastructure choices that determine whether AI systems work in production.

When the model was right but nobody believed it
When the model was right but nobody believed it
04 Aug, 2026 | 05 Mins read

An agriculture technology company built a crop yield prediction model that combined satellite imagery, soil sensor data, weather forecasts, and historical yield records. The model predicted per-field

Should every company build their own LLM? A contrarian view
Should every company build their own LLM? A contrarian view
03 Aug, 2026 | 05 Mins read

A pharmaceutical company I consulted for was three months into a project to fine-tune a large language model on their internal research corpus. The project had a team of four engineers, a budget of $8

Setting up a model registry: the minimal viable approach
Setting up a model registry: the minimal viable approach
02 Aug, 2026 | 06 Mins read

A model registry is the version control system for your trained models. Without one, teams track model versions by filename, store artifacts in ad-hoc cloud storage locations, and discover which model

AI safety regulation roundup: US, EU, UK, and Asia compared
AI safety regulation roundup: US, EU, UK, and Asia compared
01 Aug, 2026 | 04 Mins read

The regulatory landscape for AI safety has fractured along jurisdictional lines. The EU has taken a prescriptive, risk-based approach. The US has taken a sector-specific, agency-led approach. The UK h

AI Costs: The Utility Meter
AI Costs: The Utility Meter
31 Jul, 2026 | 09 Mins read

Your office building has one electricity meter. At the end of the month, you get a bill for the whole building. You know the total cost of electricity for the month. You do not know which floor consum

Schema registry showdown: Confluent vs Apicurio vs AWS Glue
Schema registry showdown: Confluent vs Apicurio vs AWS Glue
30 Jul, 2026 | 05 Mins read

When producers and consumers share a Kafka topic without agreeing on the data format, things break in production. A producer adds a field. A consumer expects the old schema. The deserialization fails,

The gender gap in AI: what the data actually shows
The gender gap in AI: what the data actually shows
29 Jul, 2026 | 05 Mins read

The headline numbers are familiar. Women represent roughly a quarter of AI and data science professionals globally. At senior levels, the proportion drops to the low teens. At the C-suite level of AI-

Building trust in AI recommendations — the change management story
Building trust in AI recommendations — the change management story
28 Jul, 2026 | 06 Mins read

A consumer goods company built an AI system that recommended reorder quantities for 12,000 SKUs across 340 distribution points. The system optimized for a multi-objective function that balanced invent

AI in the Software Development Lifecycle: From Code Review to Deployment
AI in the Software Development Lifecycle: From Code Review to Deployment
27 Jul, 2026 | 22 Mins read

Code completion gets the attention, but it is the narrowest part of what AI can do in a development workflow. Walk into any team that has shipped software for a few years and they will tell you: writi