AI Solutions Delivery Lead

About Alkem:

Alkem Laboratories Limited is an Indian multinational pharmaceutical company headquartered in Mumbai, that manufactures and sells pharmaceutical generics, formulations and nutraceuticals in India and globally over 50 countries. We have consistently been ranked amongst the top five pharmaceutical companies in India. Our portfolio includes illustrious brands like Clavam, Pan, Pan-D, and Taxim-O, which feature amongst the top 50 pharmaceutical brands in India.

About Alkem GCC:

Alkem's Global Capability Centre in Airoli, Navi Mumbai is the company's transformation engine. The GCC consolidates and runs global processes across HR Operations, IT, Sales Enablement, Procurement and Master Data for Alkem's domestic and international entities, with Finance and Supply Chain on the transition roadmap. The mandate is AI-led automation, international process transition and capability building.

Role purpose:

This role takes a requirement through build to deployment, and owns what makes the output trustworthy enough to sit in a payment path, a compliance path or a hiring decision. It owns the test libraries, the quality assurance and validation function, the release gates and the production controls.

Key responsibilities:

1.Solution Delivery

• Own the delivery lifecycle from approved requirement to production, running to the GCC three to four week cadence.
• Deliver and sustain the live and pipeline portfolio, including expense validation, background verification pre-screening, settlement processing, master data solutions, predictive analytics dashboards and the intelligence center roadmap.
• Direct the contracted AI developer pool: allocation, technical standards, code review and acceptance.
• Stay hands-on enough to read the code, the prompts and the failing test.

2.Quality Assurance, Validation and Test Libraries

• Own quality assurance and validation for all AI output. This function reports to this role.
• Maintain a behavioral test library per solution. A test must fail when the logic is wrong, not only when the code breaks.
• Apply mutation testing to control logic and treat a surviving mutation as a gap in the suite.
• Measure accuracy as precision and recall against a stated human baseline, and re-measure on every material change.

3.Release Gates and Controls

• Define and enforce the release gate. Nothing ships without a passing behavioral suite, a documented control set, a stated accuracy measure and a named business owner sign off.
• Design human in the loop placement deliberately: what the system decides, what it defers and what it escalates.
• Ensure that where automation replaces a manual control, the replacement control is designed, documented and tested.

4.Prompt and Model Governance

• Manage prompts as controlled versioned documents with changelogs that record the rationale for each change, not only the change.
• Pin model versions, and re-validate against the test library on any model or platform change.
• Maintain the reference data and configuration used by each solution, with change detection between stages.
• Operate within the Alkem Azure OpenAI tenant and the approved tooling set, and hold the boundary.

5.Production Integrity and Audit Trail

• Ensure every production run is recorded, every output is traceable to its inputs, and no sealed output is edited in place.
• Provide a governed correction and re-issue path with provenance, so that a correction supersedes rather than overwrites.
• Own incident management for AI solutions: containment, root cause, corrective action and the test that prevents recurrence.
• Produce the evidence internal audit requires for any AI-assisted control.

6.Team and Standards

• Lead the AI Business Analysts and AI Operations Analysts, set their technical standards and develop their capability.
• Work with the Lead, AI Incubation and Horizon Scouting to receive promoted proofs with a specification and a test set, and hold the quality of that handover.
• Work with the Head of GCC Excellence and Program Management on portfolio sequencing, benefit baselines and release slotting.

Candidate profile:

Experience

• 10 to 14 years, with at least four years leading delivery of AI, automation or data products into enterprise production.
• Direct ownership of a test and release discipline, not only a build backlog.
• Experience deploying AI into a process with financial, compliance or personnel consequence, where a wrong output has a cost.
• Experience managing contracted development capacity to a technical standard.
• Regulated industry exposure preferred.

Technical Skills

• Large language model applications in production: prompt engineering, retrieval, structured output, agent orchestration and evaluation.
• Azure OpenAI within a corporate tenant, and the associated security and data handling constraints.
• Python at working depth. Test frameworks, behavioral testing and mutation testing.
• RPA and workflow at scale: Power Automate, Automation Anywhere, UiPath or equivalent.
• Integration with SAP S/4HANA, SuccessFactors, SAP Ariba, Freshservice and GARNET.
• Document intelligence and extraction pipelines, including their failure modes.
• Version control, release management and configuration management practice.

Education and Certifications

• Bachelor's degree in Engineering, Computer Science or Information Technology. Postgraduate qualification preferred.
• Cloud, AI or automation platform certification is an advantage.