AI Architect

Confiz

23h

About The Position

We are seeking an exceptional Senior AI Engineer to join our team and drive the development of cutting-edge AI solutions that includes conversational, Agentic AI's. This role focuses on building advanced conversational AI systems, implementation AI Agents across different use cases, implementing sophisticated RAG architectures, and deploying production-grade voice AI applications. The ideal candidate will have deep expertise in LLM technologies and a proven track record of delivering scalable AI solutions. Key Responsibilities Design and implement advanced Retrieval-Augmented Generation (RAG) systems with sophisticated chunking strategies, multi-vector retrieval, and hybrid search architectures Develop and optimize conversational AI agents with complex multi-turn dialogue management and context awareness Build production-ready voicebot solutions integrating real-time speech processing pipelines Architect and deploy Text-to-Speech (TTS) and Speech-to-Text (STT) systems with low latency and high accuracy Design and Deploy PROD grade AI Agents that can scale up for various use cases. Fine-tune large language models using various techniques. Optimize model performance through prompt engineering, few-shot learning, and retrieval strategies Implement evaluation frameworks and metrics for LLM outputs and conversational quality Develop custom training pipelines for domain adaptation and specialized use cases Design scalable AI infrastructure supporting real-time inference and high-throughput batch processing Implement efficient vector databases and semantic search systems (Pinecone, Azure Search) Build monitoring and observability solutions for AI systems in production Stay current with latest developments in LLM architectures, RAG techniques, and voice AI technologies Experiment with emerging models and frameworks to evaluate applicability to business problems Contribute to technical documentation, best practices, and knowledge sharing within the team Prototype and validate new AI capabilities through POCs and pilots

Requirements

8+ years of software engineering experience with 4+ years focused on AI/ML engineering
Strong proficiency in Python and ML frameworks (PyTorch, TensorFlow)
Master's or PhD in Computer Science, AI/ML, or related technical field
Experience with open-source LLMs (Llama, Mistral, etc.) and local deployment
Knowledge of reinforcement learning from human feedback (RLHF) and DPO
Familiarity with multi-modal AI systems (vision-language models, audio-visual understanding)
Experience with real-time streaming architectures and WebRTC
Published research or contributions to open-source AI projects
Experience with GPU optimization and distributed training
Knowledge of voice biometrics, speaker diarization, or emotion recognition
Background in linguistics, phonetics, or speech processing
Deep hands-on experience with advanced RAG systems: Multi-stage retrieval architectures (hybrid search, re-ranking) Embedding models and vector similarity search Document processing, chunking strategies, and metadata filtering Query understanding and retrieval optimization
Multi-turn dialogue management Intent recognition and slot filling Context tracking and memory management Integration with LLM APIs (OpenAI, Anthropic, Cohere, etc.)
Production experience with voicebot development: Real-time audio streaming and processing Voice activity detection (VAD) End-to-end latency optimization Integration of STT, dialogue management, and TTS components
Expertise in TTS and STT technologies: Modern TTS systems (ElevenLabs, Soniox, Azure Speech) STT engines ( Deepgram, AssemblyAI) Audio preprocessing and enhancement Voice cloning and customization
Strong experience with LLM fine-tuning:
Proficiency with cloud platforms (AWS, GCP, or Azure) and containerization (Docker, Kubernetes)
Experience with MLOps tools and practices (MLflow, Weights & Biases, experiment tracking)
Strong understanding of API design, microservices architecture, and system integration
Knowledge of version control, CI/CD pipelines, and deployment automation

Responsibilities

Design and implement advanced Retrieval-Augmented Generation (RAG) systems with sophisticated chunking strategies, multi-vector retrieval, and hybrid search architectures
Develop and optimize conversational AI agents with complex multi-turn dialogue management and context awareness
Build production-ready voicebot solutions integrating real-time speech processing pipelines
Architect and deploy Text-to-Speech (TTS) and Speech-to-Text (STT) systems with low latency and high accuracy
Design and Deploy PROD grade AI Agents that can scale up for various use cases.
Fine-tune large language models using various techniques.
Optimize model performance through prompt engineering, few-shot learning, and retrieval strategies
Implement evaluation frameworks and metrics for LLM outputs and conversational quality
Develop custom training pipelines for domain adaptation and specialized use cases
Design scalable AI infrastructure supporting real-time inference and high-throughput batch processing
Implement efficient vector databases and semantic search systems (Pinecone, Azure Search)
Build monitoring and observability solutions for AI systems in production
Stay current with latest developments in LLM architectures, RAG techniques, and voice AI technologies
Experiment with emerging models and frameworks to evaluate applicability to business problems
Contribute to technical documentation, best practices, and knowledge sharing within the team
Prototype and validate new AI capabilities through POCs and pilots