Rebuilding the same AI plumbing?

Build the core once.
Use it across products.

Every new feature repeats the same model calls, logs and evaluation setup. Put that work in one shared service your team can extend.

Learn from production experience

I’m Ion Moșnoi.

ML engineer. Product builder. Your instructor.

8+ years in ML and software. Dozens of projects in production.

I bring the decisions behind those systems into your team’s work: what failed, how we measured it and what we changed.

ShortsBlink: one model gateway reused across applications.

LLM Arena and TennisLens: compare outputs, review mistakes, improve the next version.

What we build

Model routing, task configuration, access boundaries and failure handling.

Datasets, experiments, evals and reviewed human corrections.

Request traces, errors, latency and model cost.

Code, docs, employee training and four weekly calls, plus text support for 30 days.

Your engineers run it, evaluate it, inspect a failure and add a task. Additional integrations and enterprise requirements are scoped separately.

Tell me what your team is building and where it breaks.