Team: Product, Engineering & Data Science

Reporting to: Principal LLM Researcher

About ELSA

ELSA is a global leader in AI-powered English communication training, dedicated to transforming how people learn and speak English with confidence. Founded in 2016 and headquartered in San Francisco, we operate across the U.S., Vietnam, Portugal, Indonesia, Brazil and Japan.

Powered by proprietary speech-recognition technology and generative AI, ELSA delivers real-time, hyper-personalized feedback to help learners improve pronunciation, fluency, and overall communication effectiveness. With over 50 million learners and 1 billion hours of anonymized speech data, ELSAs depth of language training intelligence is unmatched in the industry.

Our B2B flagship platforms ELSA Enterprise and ELSA Schools empower organizations and educational institutions to elevate communication capabilities and unlock personal and professional opportunities for their people. We design engaging, bite-sized learning experiences that adapt to each learners goals and context, ensuring measurable improvement and lasting confidence.

Our vision is to become the global standard for real-time English communication training, enabling 1.5 billion language learners worldwide to speak clearly, be understood, and share their stories with the world.

Backed by world-class investors including Googles Gradient Ventures, Monks Hill Ventures, and SOSV, ELSA has been recognized among the top global AI innovators:

• Forbes Top 4 Companies Using AI to Transform the World

• Research Sniper Top 5 Best AI Apps

• ASU+GSV EdTech 150

• CB Insights Top 100 AI Companies

Join us in shaping the future of language learning and empowering millions to unlock opportunity through confident communication.

Job overview:

Join the Gen AI & LLM Team to develop and deploy LLM-powered capabilities, enhancing ELSA’s conversational tutoring. This AI Engineer role bridges software engineering and machine learning, requiring a pragmatic, change-driving individual to accelerate AI development, evaluation, and deployment. You will collaborate with researchers and engineers to implement cutting-edge AI in scalable, real-world applications.

Key Responsibilities

• Design, build, and deploy production-grade AI agents for conversational and task-oriented experiences.

• Architect scalable agentic systems (memory, tool orchestration, multi-step workflows), ensuring reliability and impact.

• Implement robust evaluation, observability, and feedback loops to drive continuous improvement in performance and cost efficiency.

• Develop secure, interoperable tool integrations (APIs, external systems, structured retrieval) adhering to modern standards.

• Collaborate with research, product, and engineering to translate AI capabilities into scalable, user-facing applications.

• Integrate speech technologies (ASR/TTS) into conversational AI systems where applicable.

What You Will Have

Must-Haves:

• Strong experience building and deploying AI systems powered by LLMs, with a focus on real-world reliability and user impact.

• Solid understanding of agentic architectures, including tool orchestration, memory design, multi-step reasoning, and structured workflows.

• Deep experience with multi-agentic memory systems, including episodic, semantic, and procedural memory design, shared memory across agents, memory consolidation strategies, and context window management in long-horizon tasks.

• Experience integrating AI systems with external APIs, structured data sources, and retrieval systems in production environments.

• Hands-on experience with MCP (Model Context Protocol) integration — including building, deploying, and managing MCP servers, tool/resource/prompt exposure patterns, and secure client-server communication in agentic pipelines.

• Hands-on experience with evaluation, monitoring, and performance optimization of AI applications (latency, cost, robustness, safety).

• Strong software engineering fundamentals, including distributed systems, APIs, containerization, and cloud-native deployment.

• Practical knowledge of prompt design, model adaptation (fine-tuning or parameter-efficient approaches), and controlled generation techniques.

• Understanding of security considerations in AI systems, including prompt injection risks, tool permission scoping, and data handling practices.

• Experience designing and orchestrating multi-agent systems with role specialization, inter-agent communication protocols, and coordination patterns (supervisor, peer-to-peer, hierarchical).

• Experience working cross-functionally with research, product, and engineering teams in fast-moving environments.

Nice-to-Have:

• Experience with multi-agent systems, planner–executor architectures, or structured reasoning frameworks.

• Familiarity with interoperability standards and tool communication protocols for agentic ecosystems.

• Experience designing long-context or memory-augmented AI systems (episodic and semantic memory strategies).

• Knowledge of advanced model optimization techniques, including parameter-efficient fine-tuning, distillation, or model compression.

• Experience integrating AI systems with speech technologies (ASR and TTS) for real-time conversational applications.

• Background in NLP, conversational AI, or education-focused AI products.

What We Offer

• Flexible work setup: Remote and hybrid work arrangements available.

• Comprehensive employee well-being benefits.

• Free ELSA Premium courses to polish your language skills

• Collaborative, international team culture.

• Opportunity to contribute to a fast-growing, well-funded Silicon Valley startup with global impact.