GenAI Architecture Developer
Key Responsibilities
System Design: Architect end-to-end GenAI solutions, including data ingestion pipelines, vector databases (e.g., Pinecone, Milvus), and model serving layers.
-
LLM Orchestration: Build and optimize complex workflows using frameworks like LangChain, LlamaIndex, or AutoGPT to create autonomous agents and multi-step reasoning systems.
-
Performance Optimization: Implement strategies for latency reduction, such as model quantization, caching, and load balancing, while managing token consumption costs.
-
Security & Governance: Establish guardrails for Prompt Injection defense, data privacy (GDPR/HIPAA compliance), and AI ethics to prevent hallucinations and bias.
-
Continuous Innovation: Stay at the forefront of the field, evaluating new foundation models (GPT-x, Claude, Gemini) and open-source alternatives (Llama 3+, Mistral).
Required Technical Skills
AI & Machine Learning
Architecture: Deep understanding of Transformers, attention mechanisms, and fine-tuning techniques (LoRA, QLoRA).
-
Frameworks: Mastery of PyTorch or TensorFlow; expert-level use of LangChain or Semantic Kernel.
-
Techniques: Proven experience with RAG, prompt chaining, and context window management.
Engineering & Cloud
Languages: Expert proficiency in Python and often a compiled language like Java or C++.
-
Infrastructure: Extensive experience with cloud platforms (GCP Vertex AI) and containerization (Docker, Kubernetes).
-
Data Systems: Proficiency in SQL, NoSQL, and Vector Databases.
Qualifications
Education: Bachelor’s or master’s in computer science, AI, or Data Science (PhD often preferred for Senior/Lead roles).
-
Experience: Typically 8+ years in IT/Software Engineering, with at least 4 -5 years specifically focused on Generative AI or Large Language Models.
