Role Overview
Own Ascend's GenAI & Agentic AI Platform — shared infrastructure enabling product teams to build, deploy, and govern LLM-powered features and AI agents at scale. You'll define platform strategy across LLM gateway, RAG pipelines, agent orchestration, prompt versioning, and AI observability.
Key Responsibilities:
Platform Strategy & Roadmap
- Own the multi-year roadmap: LLM gateway, RAG, agent orchestration, prompt management, and AI guardrails.
- Translate business and engineering needs into a prioritised backlog with clear metrics (time-to-production, developer NPS, latency, cost per inference).
- Make build-vs-buy decisions on LLM providers, vector stores, agent frameworks, and eval tooling.
- Track the GenAI landscape and assess emerging models and frameworks (LangChain, LlamaIndex, AutoGen).
Developer Experience & Adoption
- Define APIs, SDKs, and self-serve tooling so squads can integrate LLMs and deploy AI agents without rebuilding boilerplate.
- Drive platform adoption; track onboarded teams, deployed features, agent usage, and reliability KPIs.
- Maintain a developer portal with documentation, prompt libraries, and best-practice guides.
AI Feature Delivery
- Write sharp PRDs and user stories defining model inputs, outputs, fallback states, and edge cases.
- Set standards for agent memory, tool use, task decomposition, and human-in-the-loop escalation.
- Establish reliability and eval frameworks for agentic workflows, including failure detection and rollback.
Governance, Safety & Compliance
- Embed responsible-AI controls: content moderation, hallucination detection, prompt injection protection, and audit logging.
- Ensure all GenAI capabilities meet BOT, PDPA, and AMLO requirements (data residency, consent, explainability).
- Own incident response for model degradation, cost overruns, and safety events.
Cross-Functional Leadership
- Align squads across payments, lending, insurance, and customer service on GenAI platform use.
- Coordinate ML engineering, data engineering, security, legal, and business stakeholders.
- Present platform strategy, capability progress, and ROI to VP/C-suite.
Qualifications:
Required
- 5+ years in product management; 2+ years owning developer-facing or platform/infrastructure products.
- Hands-on familiarity with LLMs, RAG architectures, and prompt engineering — can assess output quality and hold technical conversations with ML engineers.
- Experience shipping products built on LLM APIs (OpenAI, Anthropic, Google Gemini, or open-source models).
- Strong analytical skills; comfortable with SQL, eval benchmarks, and cost-latency-quality trade-offs.
- Excellent written and verbal English; able to write crisp PRDs, strategy memos, and stakeholder decks.
Preferred
- Exposure to agentic AI frameworks: LangChain Agents, AutoGen, CrewAI, or similar.
- Experience with vector databases (Pinecone, Weaviate, pgvector) and RAG pipeline design.
- LLM evaluation and red-teaming / safety testing experience.
