Design and implement ingestion and transformation workflows for unstructured content (e.g., PDFs, reports, emails), including chunking, semantic tagging, and metadata enrichment.
Build and manage vector embedding services using leading LLMs (e.g., Anthropic, Meta, Mistral), and integrate with vector databases such as
OpenSearch ,
FAISS ,
AWS Kendra , or
Azure AI Search .
Develop and operationalize
RAG pipelines
using
AWS Bedrock , incorporating Knowledge Bases and GenAI agents.
Create interactive UIs using
React
or
Streamlit
to support end-user exploration and validation of GenAI results.
Implement infrastructure automation and deployment pipelines using
Terraform ,
GitHub Actions , and
ArgoCD , supporting GitOps practices across environments.
Prototype and scale GenAI solutions in partnership with product and data teams, helping define the architectural blueprint for intelligent applications.
Establish reusable frameworks and automation patterns that accelerate AI adoption and integration into business systems.
Required Qualifications
3+ years of hands-on experience
in GenAI or ML infrastructure, including:
Designing and managing vector databases (e.g.,
FAISS ,
Weaviate ,
OpenSearch )
Strong experience transforming unstructured data into AI-consumable formats, including:
Semantic chunking
Content classification
Practical experience integrating hosted LLMs (e.g., via
AWS Bedrock ,
Azure AI Foundry ) into production pipelines.
Proven success in building GenAI solutions involving:
Adapter-based fine-tuning
(e.g.,
LoRA ,
PEFT )
Deep familiarity with
CI/CD and GitOps
practices:
GitHub Actions ,
ArgoCD
Infrastructure-as-Code
using
Terraform
Kubernetes
deployments, particularly on
AWS EKS
Preferred Qualifications
Experience deploying LLM-based systems in regulated or enterprise environments.
Familiarity with multi-modal GenAI workflows (text, images, documents).
Background in MLOps, ML governance, or AI observability frameworks.
Contributions to open-source GenAI or MLOps tooling.