Responsible Agent Investment Portfolio Adviser

The mandate

The committee's standing question is which agentic AI investments justify scaling when autonomy can increase both useful work and the cost of failure. The adviser will test the portfolio's value and responsible-operating assumptions. The remit is not to select a favourite model vendor or turn exploratory demonstrations into an enterprise-wide strategy by endorsement.

Four reserved days monthly cover portfolio evidence, a targeted use-case challenge and committee attendance with a written opinion. Attendance is included. Questions are acknowledged within one working day and receive an analytical response within three where evaluation evidence is provided. Continuous architecture support and incident command are outside the retainer.

The initial term is nine months from 19 October 2026. Renewal is determined by the committee chair after reviewing whether advice improved the discipline of scaling choices. The adviser receives no line authority; executive responsibility for investment and deployment remains internal. Product, risk and technology owners retain their respective approval and operating obligations.

The sponsor supplies authorised use-case evaluations, adoption evidence, cost scenarios and risk classifications. Advice must distinguish what an agent can perform in a controlled test from what an operating team can safely rely upon. Recommendations should identify where additional autonomy creates value, where human review remains economically sensible and which failure mode would invalidate the proposed scaling case.

Concurrent independent work is allowed unless it creates a conflict with the investment opinion. Vendor remuneration, equity in a proposed solution or advice to a directly competing deployment programme must be disclosed before review. Procurement execution, formal assurance certification and operational delivery are excluded. The committee wants a practice leader's independent judgement without confusing influence with the authority to deploy.

What you will own

• Challenge scaling cases against observed adoption and failure evidence, pressing sponsors to distinguish test capability from an economically dependable operating service.
• Test autonomy assumptions for human-review cost and escalation practicality, advising where a narrower agent role may produce better value than maximum automation.
• Shape portfolio comparisons that include inference, integration and operating support, avoiding a ranking based solely on demonstration quality or nominal model cost.
• Examine evaluation coverage for consequential failure modes, identifying which missing test should prevent stronger investment confidence until evidence is obtained.
• Press the committee to stage commitments behind learning gates, making clear what must be observed before a broader deployment receives funding.
• Review proposed vendor and ecosystem dependencies for reversibility, advising where commercial concentration or unavailable evidence weakens the investment case and changes the proposed learning gate.
• Record independent advice, dissent and recusal conditions in a portfolio note, leaving architecture choice, risk acceptance and binding budgets with authorised owners.

Candidate qualifications

• Evidence AI strategy or practice leadership involving investment choices, not only solution demonstrations. Explain a portfolio decision, the value assumptions challenged and why a different autonomy or sequencing choice was recommended.
• Demonstrate responsible AI judgement grounded in operating evidence. Candidates should describe a failure mode that changed investment confidence and show how evaluation, human review and escalation were considered together.
• Provide an example of comparing AI economics across use cases with different integration and support burdens. Explain the cost that was initially omitted and the financial consequence of making it visible.
• Show advisory independence from vendor and investment incentives. Describe a disclosed conflict, a recusal or a recommendation that resisted a sponsor's request for endorsement, keeping executive deployment authority clearly separate.
• Be able to sustain the four-day monthly rhythm and deliver usable opinions under uncertainty. Demonstrate enterprise investment judgement through a consequential AI use case, confidential handling of its evaluation records and a clear distinction between commercial advice and operating assurance. Show an investment opinion where evaluation limits prevented a confident return comparison. Explain how you framed a useful next learning decision, which evidence would change the ranking and how the committee understood that uncertainty without receiving an empty recommendation to investigate further.

Application

Applications for this mandate are received in one way only: through the India Board Terminal's application process. It is automated end to end. Your Executive Passport travels to the mandate holder in its confidential form, your answers to the three questions below are read before anything else in your file, and every stage that follows is recorded on your applications page.

There is no address to write to and no intermediary to call. The mandate holder reads what the Terminal delivers and nothing else, which is what keeps the process the same for every applicant and keeps your name out of it until you release it. Applications close on 11 October 2026. Mandate reference PCT-ADV-2026-IND-10.