analytics practice · Cohort
Model Evaluation Studio
Blend qualitative rubrics with lightweight quantitative checks. You will calibrate human review loops, annotate edge cases, and narrate results to non-technical sponsors without overselling certainty.
4 weeks Intermediate Afternoons ICT
11,200 THB
Informational pricing—see Returns & Refunds for eligibility.
Request informationInside the syllabus
- Rubric design sprint
- Calibration games between annotators
- Scorecards that plug into Looker Studio
- Synthetic data cautions module
- Bias note templates
- Live critique of two real model cards
- Documentation pack for procurement teams
Outcomes you can show
- Publish a model card draft with honest limitations called out
- Run a two-pass annotation round with measured agreement
- Present findings using the studio's executive one-pager format
Mira Langston
Evaluation lead for civic-tech pilots across ASEAN cities.
Straight FAQ
Math depth?
We stay at descriptive statistics; no calculus required.
Team tickets?
Teams of three may share a submission; pricing does not drop automatically—email us for volume notes.
What is out of scope?
We do not certify you for external audits—this is practice, not legal sign-off.
Recent notes
Model card workshop forced us to list what we *won't* claim—uncomfortable but overdue.
Tik · PM · Fintech squad · Trustpilot