You will design, build, and deploy production-grade AI systems — including LLM-powered conversational agents, RAG pipelines, NLP workflows, and voice AI integrations — to deliver intelligent, reliable, and measurable AI solutions for enterprise clients across government, financial services, healthcare, and telecommunications sectors — so that Kata's clients can automate customer interactions at scale with high accuracy, low latency, and strong business impact.

Qualifications & Education :

Bachelor's degree in Computer Science, Artificial Intelligence, Data Science, Computational Linguistics, or related field
Master's degree in AI/ML is a plus
Relevant certifications (GCP AI/ML, DeepLearning.AI, etc.) are advantageous

Technical Skills :

LLM Integration: OpenAI GPT-4o, Anthropic Claude, Google Gemini, or open-source models (LLaMA, Mistral, Qwen)
AI Frameworks: LangChain, LlamaIndex, CrewAI, or similar agent/RAG orchestration frameworks
Prompt Engineering: System prompt design, few-shot prompting, chain-of-thought, structured output (JSON mode, function calling)
RAG Pipelines: Document chunking, embedding strategies, retrieval optimization, reranking
Vector Databases: Pinecone, Weaviate, Qdrant, or pgvector
Voice AI: LiveKit Agents SDK, STT integrations (Deepgram, Google Speech-to-Text, Whisper), TTS integrations (ElevenLabs, Google TTS)
Languages: Python (required); FastAPI for AI service exposure
Cloud: GCP or Azure for AI/ML workload deployment — Vertex AI, Azure OpenAI, Cloud Run
Evaluation Frameworks: RAGAS, DeepEval, custom eval pipelines, or LLM-as-judge approaches
Containerization: Docker; basic Kubernetes for deploying AI services
Monitoring: AI-specific observability — LangSmith, Langfuse, or custom logging for tracing LLM calls in production

Skills & Competencies :

Strong analytical thinking — able to diagnose and improve AI system behavior through systematic evaluation and iteration
Ability to balance research-oriented experimentation with production-grade engineering rigor
Good communication skills — able to explain AI system behavior, limitations, and tradeoffs clearly to non-technical Product and Project stakeholders
Quality-first mindset — proactively defines success metrics and failure modes before building
Collaborative and cross-functional — works closely with Backend, Frontend, and QA to ensure AI components integrate cleanly into the full product
Agile/Scrum mindset with comfort operating in sprint-based delivery cycles
Keeps up with the fast-moving AI landscape and proactively identifies relevant new tools, models, or techniques applicable to the team

AI Engineer

TLDR