Technology Architect

Job Summary

Gen AI Agentic AI Project Management

Python

Designing scalable GenAI systems (e.g. RAG pipelines multi-agent systems).

Choosing between hosted APIs vs open-source models.

Architecting hybrid systems (LLMs + traditional software).

• Model Evaluation & Selection
  
  

Benchmarking models (e.g. GPT-4 Claude Mistral LLaMA).

Responsibilities

Strategic & Leadership-Level GenAI Skills

• AI Solution Architecture
  
  

Designing scalable GenAI systems (e.g. RAG pipelines multi-agent systems).

Choosing between hosted APIs vs open-source models.

Architecting hybrid systems (LLMs + traditional software).

• Model Evaluation & Selection
  
  

Benchmarking models (e.g. GPT-4 Claude Mistral LLaMA).

• Understanding trade-offs: latency cost accuracy context length.
  
  

Using tools like LM Evaluation Harness OpenLLM Leaderboard etc.

• Enterprise-Grade RAG Systems
  
  

Designing Retrieval-Augmented Generation pipelines.

Using vector databases (Pinecone Weaviate Qdrant) with LangChain or LlamaIndex.

Optimizing chunking embedding strategies and retrieval quality.

• Security Privacy & Governance
  
  

Implementing data privacy access control and audit logging.

• Understanding risks: prompt injection data leakage model misuse.
  
  

Aligning with frameworks like NIST AI RMF EU AI Act or ISO/IEC 42001.

• Cost Optimization & Monitoring
  
  

Estimating and managing GenAI inference costs.

Using observability tools (e.g. Arize WhyLabs PromptLayer).

Token usage tracking and prompt optimization.

Advanced Technical Skills

• Model Fine-Tuning & Distillation
  
  

Fine-tuning open-source models using PEFT LoRA QLoRA.

Knowledge distillation for smaller faster models.

Using tools like Hugging Face Axolotl or DeepSpeed.

• Multi-Agent Systems
  
  

Designing agent workflows (e.g. AutoGen CrewAI LangGraph).

Task decomposition memory and tool orchestration.

• Toolformer & Function Calling
  
  

Integrating LLMs with external tools APIs and databases.

Designing tool-use schemas and managing tool routing.

Team & Product Leadership

• GenAI Product Thinking
  
  

Identifying use cases with high ROI.

Balancing feasibility desirability and viability.

Leading GenAI PoCs and MVPs.

• Mentoring & Upskilling Teams
  
  

Training developers on prompt engineering LangChain etc.

Establishing GenAI best practices and code reviews.

Leading internal hackathons or innovation sprints.

---