Artificial Intelligence Engineer

Core Must-Have Requirements (Non-Negotiable)

• Agentic AI Experience Built AI agents in production.
• Experience with memory systems, RAG, retrieval pipelines, tool calling, workflow orchestration.
• Worked with frameworks such as LangGraph, LangChain, CrewAI, AutoGen, Semantic Kernel, etc.
• Experience designing multi-step agent workflows rather than simple chatbot integrations.
• Self-Hosted Model Deployment Hands-on experience deploying and serving open-source models.
• Experience with:
• vLLM
• Triton Inference Server
• Ollama (less preferred)
• TGI (Text Generation Inference)
• Strong understanding of GPU utilization, batching, latency optimization, throughput, and inference costs.
• LLM Gateway & Orchestration Experience managing multiple model providers and routing strategies.
• Exposure to:
• LiteLLM
• Portkey
• OpenRouter
• Custom LLM gateways
• Model selection, fallback logic, observability, rate limiting, and cost controls.

Strongly Preferred

• Production AI Infrastructure Kubernetes
• Docker
• GCP (preferred) or AWS
• CI/CD for ML systems
• Monitoring and alerting
• Evaluation & Observability Building evaluation frameworks
• A/B testing models
• Tracking metrics such as:
• WER (speech)
• DER (diarization)
• Recall/F1 (retrieval)
• Latency
• Cost per request
• Experience with Langfuse, Arize, Weights & Biases, OpenTelemetry, Grafana, Prometheus, etc.

Good-to-Have

• Speech/Audio AI Whisper, Deepgram, AssemblyAI, Speechmatics
• Speaker diarization
• Voice identification
• Real-time audio pipelines
• Research Background Publications
• Applied ML research
• Fine-tuning models

Ideal Candidate Background

• 3–5 years of AI/ML Engineering experience.
• Strong Python engineer.
• Has built and shipped agentic AI systems.
• Has deployed and optimized self-hosted LLMs.
• Understands model serving, observability, evaluation, and infrastructure.
• Audio/STT experience is a bonus, not mandatory.