Senior AI/ML Engineer Agentic Systems

Experience: 4–8 years | 2+ years with LLMs in production

Location: Office – Coimbatore/Bengaluru

Aivar is an AI -first technology partner where cutting -edge technology meets industry expertise to supercharge your projects. Our AI -augmented teams accelerate development, reduce time -to -market, and deliver exceptional code quality. We bring together the best minds in tech to craft scalable, repeatable solutions that drive real momentum for your business.

Own accelerators’ intelligence core: the governance processing engine, multi -agent document extraction, knowledge graph construction, LLM orchestration, and the Evals Engine for cross -model benchmarking. Ship production systems processing 100K+ enterprise documents.

• Build the governance engine — reasoning against business rules, compliance validation with green/red -line thresholds, and action execution
  
• Design multi -agent orchestration across the 6 -n stage pipeline with structured reasoning chains and tool -use patterns
  
• Implement multi -format document extraction (PDF, TIFF, JPG, structured data) using OCR, layout analysis, and table extraction models
  
• Build the decision graph / context graph that accumulates institutional knowledge from human -in -the -loop interventions
  
• Develop the Evals Engine for A/B testing across LLM providers (GPT -4, Claude, Llama, Mistral) with regression testing
  
• Implement responsible AI guardrails — bias detection, confidence calibration, explainability layers
  

• Production LLM experience —Anthropic, Bedrock APIs + open -source models (Llama, Mistral)
  
• Agentic frameworks — Strands Agents, LangChain, LangGraph, CrewAI, AutoGen, or custom multi -agent systems
  
• Document AI — OCR, layout parsing (LayoutLM), table extraction, Textract/Unstructured.io
  
• Vector databases — OpenSearch, Pinecone, Weaviate, pgvector; RAG architecture design
  
• Strong async Python engineering — production -grade, not notebook -level
  
• AWS ML services — SageMaker, Bedrock, Lambda, S3, EKS
  
• Model evaluation methodologies — accuracy metrics, A/B testing, LLM regression suites
  

Python, FastAPI, LangChain/LangGraph, Ray, OpenAI/Anthropic/Bedrock APIs, Hugging Face, Pinecone/Weaviate/pgvector, Tesseract/LayoutLM/Textract, Prometheus/Grafana, AWS (SageMaker, Bedrock, Lambda, S3)

Benefits