Senior AI Engineer

About the job

Do you think of yourself as a modern-day artist? Read on. About the role: We’re hiring a hands-on Senior AI Engineer to build and ship practical AI features (LLM-based apps, RAG pipelines, search/embeddings, and real-time integrations). This role blends backend engineering, ML systems, and production deployment — ideal for someone who turns research ideas into reliable products. Responsibilities: Design and implement LLM-driven features and Retrieval-Augmented Generation (RAG) pipelines. Ingest and preprocess documents (PDFs, HTML, etc.), build embeddings, and implement vector search. Integrate models (cloud or local) for generation and embeddings; ensure responses are grounded with source citations. Build APIs and microservices (Node.js/Python) for inference, streaming, and real-time use cases. Implement guardrails: prevent hallucinations, handle out-of-scope queries, and protect against prompt injection. Deploy, monitor, and optimize AI services on cloud infra (AWS preferred): cost, latency, and autoscaling. Collaborate with frontend engineers to integrate chat/voice interfaces (WebSockets / WebRTC). Write clean, testable code and maintain documentation and runbooks. Required skills: 3+ years professional experience in AI/ML engineering or backend engineering with AI products. Practical experience with LLMs, embeddings, and RAG-style retrieval systems. Experience with at least one vector DB or library (e.g., MongoDB Vector Search, Chroma, FAISS, Pinecone). Strong backend skills in Node.js or Python (APIs, queues, workers). Familiarity with deploying services on AWS (EC2/Lambda/S3); basic infra & monitoring knowledge. Knowledge of WebSockets/SSE and basic WebRTC concepts for real-time/voice use cases. Good software engineering practices: testing, logging, versioning, and documentation. Clear communicator who can explain technical tradeoffs to engineers and non-engineers. Nice-to-have: Experience with streaming LLM outputs (token-by-token), hybrid search (BM25 + embeddings), or agent/tool-calling architectures. Background with ASR/TTS (Whisper or similar), model fine-tuning, or local LLM deployments. Familiarity with security/privacy concerns for embeddings and user data.

Requirements

• AI/ML
• LLM
• Node.js
• Python
• embeddings
• RAG

Preferred Technologies

• AI/ML
• LLM
• Node.js
• Python
• embeddings
• RAG

Benefits

• work from home on select days
• mental health support
• winter break
• extended long weekends
• caregiver leaves
• learning initiatives
• performance appreciation