Software Engineer, Artificial Intelligence/LLM (Multiple Seniority Levels)

Tasks

• Add access controls and auditing
• Add streaming backoff and caching to LLM services
• Add tracing logs and dashboards for model calls and token usage
• Build and maintain Python or TypeScript APIs and workers
• Create offline evals and golden sets
• Create schema bound JSON outputs with validation retries and fallbacks
• Debug failures across retrieval prompts tools and providers
• Design human in the loop paths
• Design retrieval augmented generation flows
• Develop chunking embeddings and indexing pipelines
• Implement content and policy checks PII detection and redaction
• Implement tool calling flows
• Integrate with AWS Bedrock OpenAI and Anthropic
• Run A B tests and prompt version rollouts with guardrails and canaries
• Set up online metrics for quality and latency
• Ship LLM powered product features end to end
• Tune vector backends for retrieval performance

Perks/Benefits

• 401k plan
• Healthcare coverage
• Hybrid work schedule
• Paid Company Holidays
• Paid time off
• Stipends

Skills/Tech-stack

A/B | A/B Testing | AWS Bedrock | Amazon Aurora | Anthropic | B testing | Backoff | Caching | Cost Optimization | Dashboards | DynamoDB | Embeddings | Guardrails | Human-in-the-loop | JSON | JavaScript | Langchain | Latency optimization | Logging | Monitoring | OpenAI | OpenAI API | OpenSearch | PII Detection | Python | RAG | Retrieval-Augmented Generation | Retry logic | Rédaction | Schema Validation | Streaming | The Loop | Tool-Calling | Tracing | TypeScript | Vector Database | Vector Search