• Design and run evaluations for LLM-based agents (quality, reliability, safety) • Analyze agent behavior and identify failure modes • Propose improvements based on empirical results • Build and enhance components such as tools, skills, memory systems, and MCP integrations • Improve orchestration, observability, and developer experience • Explore approaches combining LLMs with rules, knowledge graphs, or symbolic reasoning • Work on use cases requiring auditability and explainability (e.g., claims, policy logic) • Help bridge probabilistic AI with deterministic systems • Partner closely with internal stakeholders or pilot users • Rapidly prototype and iterate on solutions in real-world contexts • Help define patterns for scaling adoption of agentic systems

• Analyzing deep learning networks and developing compiler optimization algorithms • Strong programming skills in CUDA including analyzing and debugging performance bottlenecks on GPUs • Defining public APIs, performance optimizations and analysis • Crafting and implementing compiler techniques for AI workloads and future NVIDIA GPUs

• Define and maintain a unified AI strategy across three lanes: internal workflow and operations, consulting engagements, and product development. • Drive alignment on goals, sequencing investments, and resolving the tensions that come with fast-moving capability building across multiple teams. • Serve as a deep resource to account leaders in client conversations, being comfortable with method choices, tradeoffs, and risk. • Own the business case for AI investments, including build cost assumptions, ROI frameworks, and tracking time and revenue return against projections. • Set the AI communication narrative and go-to-market positioning, aligned to Beghou's overall GTM strategy. • Define how to staff and deliver against custom AI and analytics projects, including assessing, shaping, and redeploying talent. • Own a centralized point of view on what our people need to work effectively in an AI-enabled environment.

• AI clinical tool validation and LLM strengthening • Lead pre-deployment validation • Produce model cards/transparency (intended use, limits, monitoring, rollback). • Partner on data quality (provenance, representativeness, refresh cadence, privacy). • Build LLM evaluation datasets/gold standards; run clinical red-teams and prompt testing; refine prompts/guardrails; assess RAG fidelity. • Clinical-in-the-loop enablement: Train clinicians to oversee AI tools and adjudicate exceptions; create SOPs/checklists/escalations for LLM use. • Collaborate with engineering and product, security, compliance, legal, care and service line leaders. • AI governance and policy: Administer clinical AI governance (ethics, safety, regulatory, privacy, change control, documentation) in partnership with the Patient Safety Officer • Define clinical use criteria and human-in-the-loop guardrails; • Maintain governance artifacts (charter, SOPs, decision logs) and audit readiness. • Quality and patient safety: Support clinical QA; URAC/NCQA activities; policy/standards/rubrics management. • Lead/participate in event detection, RCA/FMEA, CAPA; education (CME/CE, targeted training). • Support client/payer audits, RFPs, VBC metrics (HEDIS, MIPS), implementations, and escalations, grievances/appeals. • Lead and/or support cross-functional efforts to analyze clinical quality performance data, identify improvement opportunities, and translate value-based care insights into health plan design enhancements that optimize outcomes, member experience, and cost efficiency. • Conduct ongoing assessments of clinical services, applying continuous improvement and evidence-based methodologies to ensure service effectiveness, measurable patient outcomes, and alignment with organizational quality standards.