over 1 year ago
Santa Clara, CA, USAMid Level
Base Salary
$150k - $400k/yr
Responsibilities
- Own and evolve the core dialog and policy engine, including agent state representation, decision logic, safety policies, and execution-layer guardrails.
- Design and maintain distributed context and memory systems with low-latency conversational-history access and integrated RAG and vector retrieval pipelines.
- Architect and deliver agentic orchestration patterns including planner-executor schemes, ReAct loops, and resilient multi-step workflows combining tools, LLMs, and stateful memory.
- Build internal SDKs and frameworks with composable graphs, execution chains, clear APIs, and type safety for reuse by product engineering and low-code teams.
- Own voice runtime components for streaming audio, barge-in detection, and turn-taking under strict latency, jitter, and QoS requirements.
- Architect secure tooling and integration infrastructure covering tool registration, authentication and authorization, rate limiting, circuit breaking, retries, and typed I/O.
- Define and monitor platform SLIs and SLOs, improve reliability and debuggability, and drive performance optimization.
- Design clear public APIs and internal abstractions while leading cross-functional technical initiatives and architectural decisions.
- Prototype advanced capabilities involving on-device personalization, privacy-preserving federated learning signals, and policy adaptation techniques.
Requirements
- 3+ years of hands-on backend engineering and distributed-systems experience, including ownership of core platforms or frameworks used by other engineering teams.
- Hands-on experience architecting, building, or operating production-grade agentic systems involving LLM calls, complex tool interactions, and stateful workflows.
- Strong knowledge of orchestration frameworks such as LangChain, LlamaIndex, or internal equivalents, plus RAG, ReAct, and multi-step planning patterns.
- Deep practical understanding of distributed systems, concurrent programming, and reliable multi-tenant cloud systems with defined latency and cost constraints.
- Proven experience designing, implementing, and rolling out frameworks or libraries adopted by internal engineering teams.
- Experience with security-sensitive systems, including authentication and authorization, isolation boundaries, data protection, and centralized policy or safety infrastructure.
- Strong technical communication skills and the ability to lead complex cross-functional technical initiatives.
- Preferred experience includes conversational AI platforms, agent frameworks, high-throughput workflow engines, real-time media systems, high-stakes enterprise platforms, federated learning, on-device personalization, or bandit-style adaptive policy systems.
Tech Stack
Categories
About Boson AI
Boson AI develops artificial intelligence systems focused on storytelling, knowledge acquisition, and insight generation for individuals and organizations. The company operates in-house GPU server infrastructure (e.g., Supermicro-based clusters) to train and serve its models, with teams handling hardware deployment and reliability. Founded in 2023 and headquartered in Santa Clara, it is privately held and raised a seed round in 2023.