Assembly-and-Test Yield Board Challenger — Semiconductors

The mandate

Yield recovery reports aggregate unlike packages, devices and test insertions, so mix change can appear as physical improvement. Engineering teams debate whether excursions originate in incoming wafer, mould compound, wire bond, equipment chamber, handler or test programme. The board’s recurring question is whether capital and customer promises rest on causal evidence or a favourable movement in portfolio averages.

The adviser commits four days monthly: one reviewing lot-level evidence, two challenging factory and product teams, and one preparing or attending the committee. Five meetings and six yield reviews are included. A written response to a material excursion or urgent tool request will be returned within two business days, without directing production or customer disposition.

The appointment lasts nine months through the next capacity and technology review. One three-month renewal may be approved if a newly introduced package creates a material mechanism not represented in the reviewed evidence. The committee decides after refreshed conflicts; monitoring improvement actions or supporting routine customer reviews cannot extend the engagement.

The adviser has no line authority and carries no executive or technical responsibility for lot release, process recipe, equipment, test limits, customer communication, capital or quality disposition. Management owns recovery; technical authorities approve changes; the board allocates investment. The challenger may press methods, comparability and causal claims but cannot hold lots or change programmes.

Current work for rival OSATs, integrated device manufacturers, equipment makers, materials suppliers, test houses, customers or semiconductor investors must be disclosed. The scope excludes recipe development, failure-analysis sign-off, product qualification, supplier selection, customer negotiation, patent advice, capital approval and performance evaluation of named engineers or executives.

Why the board wants this voice

Directors see sophisticated charts but lack a former assembly-and-test operator who can detect mix effects, weak containment and untested physical mechanisms. Independent challenge will improve capital and customer decisions while preserving the factory’s technical, quality and commercial authority. This perspective will connect recurrence patterns to future customer-commitment risk.

What you will own

• Press teams to stratify yield by device, package, process step, equipment, material, test insertion and time.
• Test causal claims against signatures, split lots, recurrence, containment escape and deliberately changed conditions.
• Challenge capital requests where recipe discipline, maintenance, material control or test correlation remains unresolved.
• Examine customer-risk statements for affected population, detection limit, traceability and continuing escape exposure.
• Compare factories objectively using common loss definitions without erasing product and technology differences.
• Maintain a committee ledger of mechanisms, evidence gaps, containment promises, expiry dates and verified outcomes.
• Probe six reviews spanning incoming wafer, bond, mould, chamber, handler and test-program hypotheses.

Candidate qualifications

• Held senior assembly, packaging, sort or final-test operations authority within semiconductor manufacturing.
• Recovered complex yield excursions using lot-level signatures, physical mechanisms, controlled split-lot evidence and verified containment.
• Distinguished mix-driven movement from true process improvement across multiple packages and product generations.
• Challenged equipment investment where operating discipline, materials or test correlation was the stronger cause.
• Advised governance bodies without displacing process, quality, customer or factory decision authority.
• Maintained documented continuing independence from OSATs, device companies, toolmakers, materials vendors and semiconductor investors globally.

Non-negotiables

• Available four days monthly for Penang work, five meetings and six yield reviews.
• Direct semiconductor assembly-and-test leadership is required; front-end fabrication alone is insufficient.
• Will disclose OSAT, device, equipment, materials, customer, test-house and investment interests.
• Will not release lots, change recipes, set test limits, select tools or contact customers.

1. 49 words maximum. Describe a semiconductor yield improvement that disappeared after product mix was controlled.
2. 49 words maximum. Which OSAT, device, equipment or materials interests would require disclosure?
3. 49 words maximum. How would you challenge a tool request without assuming process-engineering authority?