# Lead – AI Optimization — Franklin Templeton

- Location: San Ramon, CA
- Posted pay: $156k–$183k/yr
- Posted: Oct 23, 2026
- Status: open (checked Nov 21, 2026)
- Apply (employer's site): https://careers.franklintempleton.com/us/en/job/FTGFTMUS869753EXTERNALENUS
- Employer record: https://visajobs.io/company/franklin-templeton-companies.md

## Sponsorship history
- Tier: Good history
- Occupation (SOC): Software Developers (15-1252)
- Employer's certified LCAs for this occupation in CA, FY2024–FY2026: 2
- …in the whole US: 10
- With this exact title: 0
- Median DOL prevailing wage, level I / II: $135k / $162k

## Description
At Franklin Templeton, we believe success is built through powerful partnerships. As a forward thinking asset manager, we build dynamic relationships with clients, understand their goals, and navigate complex markets together. We leverage cutting edge strategies and deep insights to unlock opportunities for long term wealth creation. Our talented, global teams bring expertise that is both broad and unique.
From our welcoming, inclusive, and supportive culture to our globally diverse business, we offer opportunities not only to help you reach your potential, but also to contribute to our clients’ success.
About the Department
This role sits within Franklin Templeton's AI Platform team, which develops and supports shared AI tools and capabilities across investment, distribution, and corporate functions. The team partners with business stakeholders, product owners, engineers, and data and governance specialists to deliver practical AI solutions responsibly.
The team is hands-on and cross-functional, with a collaborative approach to solving business and technical problems. Team members are expected to challenge assumptions constructively, make sound engineering decisions, share knowledge, and take ownership of delivering solutions that are useful, reliable, and maintainable.
Job Overview
The Lead - AI Optimization drives the technical delivery and continuous improvement of AI solutions across business groups. Working with engineers, data scientists, product owners, and business stakeholders, the role translates business needs into practical technical approaches and guides initiatives from discovery and solution design through implementation, testing, and production readiness.
Evaluation is a core responsibility: the role leads the development and improvement of reusable evaluation capabilities for large language models (LLMs) and AI applications. It establishes meaningful benchmarks, helps teams understand performance and limitations, and uses the findings to guide model selection and improvements in quality, reliability, latency, and cost.
This is a hands-on individual-contributor role with no people-management responsibilities. Leadership is exercised through technical ownership, direct contribution, mentoring, and influence. Clear communication is essential to aligning teams, explaining technical trade-offs, and keeping stakeholders informed of decisions, progress, and risks.
How you will add value
Lead technical delivery: Translate business needs into practical designs, assessing feasibility, dependencies, and AI-assisted versus traditional approaches. Build critical components, review code, and guide initiatives through testing, deployment, and production readiness.
Own the evaluation framework: Develop reusable tools for LLM and application evaluation, including benchmarks, representative datasets, scoring methods, and regression tests. Combine automated checks, LLM-as-judge methods, and human review to assess accuracy, grounding, safety, and reliability.
Optimize quality and performance: Apply advanced prompting and improve retrieval, context construction, model selection, routing, and serving approaches. Use appropriate tuning, caching, and batching to improve quality and throughput while reducing latency and unit cost.
Validate improvements with evidence: Run structured experiments and A/B tests with clear hypotheses, guardrails, and rollback criteria. Compare models, techniques, and vendors against internal benchmarks, document limitations, and use findings to prioritize changes.
Align teams and communicate clearly: Partner across business and technical teams to resolve dependencies and meet enterprise standards. Explain trade-offs and evaluation results, document decisions, and communicate progress, risks, and next steps.
Scale effective practices: Mentor engineers and share reusable components, evaluation tooling, and practical guidance. Apply AI where it improves testing, code review, documentation, and monitoring, and measure the benefits.
Technical Environment
Core stack: Python, LLM and machine-learning frameworks, evaluation and experimentation libraries, and retrieval and vector tooling for AI applications.
Platforms: Cloud or hybrid model platforms, approved LLM endpoints, model gateways and registries, and GPU or accelerated compute where required by the solution.
Engineering workflow: Version-controlled code and experiments, automated tests, reproducible evaluations, continuous integration, and staged deployment.
Quality and reliability: Application and model monitoring, quality benchmarks, latency and cost metrics, regression detection, safety guardrails, and feedback from production use.
What will help you be successful in this role
Bachelor's degree in Computer Science, Engineering, or a quantitative discipline, or equivalent practical experience, and six or more years in applied machine learning or AI engineering, including production LLM or model-serving systems. An advanced degree is preferred.
Demonstrated technical leadership from ambiguous requirements through delivery, with the ability to align cross-functional partners, mentor engineers, and balance new initiatives with ongoing optimization without formal management authority.
Strong hands-on Python development and software-engineering judgement, including solution design, code review, testing, troubleshooting, and building maintainable production components.
Hands-on experience evaluating LLMs and AI applications using representative datasets, scoring methods, automated tests, and failure analysis. Familiarity with LLM-as-judge methods and human evaluation, including validating results against expert judgement and using findings to compare models and identify regressions.
Strong experience in prompt and retrieval optimization and model performance tuning, with practical evidence of improving quality, latency, reliability, and cost in production systems.
Rigorous experimental discipline and statistical judgeme

> History, not a promise: past H-1B filings do not mean an employer will sponsor you or a given role. Data: U.S. DOL OFLC LCA disclosure (FY2024–FY2026 for role counts, certified LCAs) and the USCIS H-1B Employer Data Hub. Not legal advice. Methodology: https://visajobs.io/methodology
