Leads customer implementations of LLM and generative-AI systems from architecture through production, building full-stack solutions and improving model quality.
About this role
This engineer joins the LTM Factory development team within the Network Systems Development organization. The team develops AI agents and business applications based on SoftBank’s Large Telecom Model, a domain-specific LLM for telecommunications, and applies them to network operations and other internal workflows. Its scope includes systems that use operational logs and evaluation data to improve agent behavior over time.
The role covers designing, coding, and implementing multi-agent systems and AI workflows; integrating LLMs, RAG, vector databases, external APIs, and tools; and rapidly prototyping applications from requirements identified by system-planning teams. It also includes building autonomous learning loops and LLMOps processes using execution logs, evaluation data, and user feedback, with development supported by tools such as GitHub Copilot.
Applicants must have experience developing web applications, AI agents, or RAG systems using generative AI or LLMs, with technologies such as Python or TypeScript. The posting also requires the ability to drive work from requirements definition and design through implementation, testing, and improvement, and to read English technical documentation and open-source repositories for product adoption.
This is an AI-generated summary of the employer's original posting — details can be incomplete, out of date or simply wrong. Always confirm everything on the official posting before applying.
Similar jobs
Leads enterprise AI implementations from requirements to production, designing architecture, data connectivity, and RAG pipelines while guiding technical standards in a hybrid Tokyo role.
Plans and develops EMS services for homes and commercial users, integrating batteries and solar devices with web/mobile apps, cloud systems, APIs, and AI-based demand and generation forecasting.
Build data pipelines and machine-learning models for commercial real estate, tune inference, and lead AI product PoCs and MVPs with business teams. Hybrid remote work is available in Tokyo.
The role involves various responsibilities with potential for transfers and business trips across locations.
Lead multiple product teams for Fanicon, aligning business, product, and technology. Manage engineers, support technical decisions, and improve processes, quality, and team growth.