Software Engineer III -backend

Suki AI Suki AI · Vertical AI · Suki, India · Engineering

Software Engineer III - Backend role at Suki AI, focusing on building and owning scalable distributed services for an AI voice solution in healthcare. The role involves implementing LLM-powered features like RAG, tool/function calling, and agentic workflows, ensuring reliability, latency, cost, and safety in production.

What you'd actually do

  1. Collaborating with the product, design and operations teams to build a world-class product
  2. Developing, owning and maintaining highly scalable distributed services and solutions
  3. Implementing a microservices-based architecture using Golang and Python
  4. Ensure application performance, uptime, and maintain high standards of code quality
  5. Build and integrate LLM-powered features — retrieval-augmented generation, tool/function calling, and agentic workflows — with care for reliability, latency, cost, and safety

Skills

Required

  • Golang
  • Python
  • distributed systems
  • microservices
  • Linux
  • data structures
  • algorithms
  • concurrent programming

Nice to have

  • gRPC
  • gRPC context
  • Protocol Buffers
  • infrastructure and systems engineering
  • GCP
  • prompting strategies
  • RAG
  • agents and tool use
  • evaluation
  • productionizing AI features
  • AI-assisted coding

What the JD emphasized

  • AI voice solutions for healthcare
  • AI assistant that uses generative AI to automatically create clinical documentation
  • AI and speech platform
  • AI is at the heart of what we build
  • LLM-powered capabilities
  • retrieval-augmented generation, tool/function calling, and agentic workflows
  • AI-fluent
  • Hands-on experience building with LLMs — prompting strategies, RAG, agents and tool use, evaluation, or productionizing AI features

Other signals

  • AI voice solutions for healthcare
  • generative AI to automatically create clinical documentation
  • ambiently listening to patient-clinician conversations
  • LLM-powered capabilities to life in production
  • retrieval-augmented generation, tool/function calling, and agentic workflows