execution-plan/weeks/week-07-measured-extraction-agent/commitment.md

execution-plan/weeks/week-07-measured-extraction-agent/commitment.md Β· UPDATED 2026-09-16

Week 07 commitment

3 parts426 words 6 checkboxes β€” tick them, they save to the file

State: UNCONFIRMED. Owner: Peter Ciancaglini. Specification

Start date: not set. End date: not set. Available hours: confirm (planned ceiling 10). Primary outcome: measured extraction agent. Minimum that keeps the next gate alive: the F1 table, whatever it says.

01

01

Deliverables to commit

  • extraction-runbook.md: Invalid JSON, absent values and unsupported evidence fail visibly.
  • evaluation-protocol.md: Each field has counts and scoring rules; development scores are not called held-out performance.
  • error-analysis.md: Unchanged held-out data and a backend decision with quality/cost evidence are recorded.
  • Bounded learning exercise and personal teach-back.
  • Evidence-backed branding and opportunity pack.
  • Sunday review and next physical action.
↑ Back to top
02

02

Proposed calendar

Day Minutes Physical action / outcome Actual slot Done
Mon 90 Implement schema-constrained extraction with provenance and fake/API/local adapters; smoke-test five development papers and log model/prompt versions. Not set [ ]
Tue 90 Michaelis-Menten fitted in Python to a public dataset Not set [ ]
Wed 90 Implement field-specific precision/recall/F1, numeric unit/tolerance checks and provenance correctness on frozen paper-level splits. Last ten minutes: optional note. Not set [ ]
Thu 90 A simple Bayesian regression in PyMC or numpyro on titer data Include relevant founder-paper work within this block. Not set [ ]
Fri 90 Inspect five worst fields, make one development-only revision, compare before/after, and choose a documented per-field floor and cost budget for scale-up. Not set [ ]
Sat 90 Resolve critical spillover; long-form draft if due; sandbox only if done. Not set [ ]
Sun 60 Build log, evidence, gap review, next actions. Not set [ ]
↑ Back to top
03

03

Dependency and capacity check

  • Missing essential inputs: not yet inspected.
  • Human/external dependency, owner and expected date: not yet inspected.
  • Week-specific constraint: The original gold set cannot be repeatedly tuned on and still serve as an untouched test set. Report prompt-development and final evaluation sets separately.
  • Existing session authorization for external actions or spending: not assessed; local planning only at scaffold creation.
  • Scope to cut first: optional polish/additional examples; preserve scientific validity.
  • If the forecast exceeds available time: record a concrete reduced scope or revised delivery proposal before committing.

Peter's commitment / adjustments: pending. Reviewing model and date: pending.

↑ Back to top
⏱ Focus Focus25:00 0 min today