Technology Architect
Bengaluru
Job No. atci-5266775-s1933382
Full-time
工作描述
Project Role : Technology Architect
Project Role Description : Design and deliver technology architecture for a platform, product, or engagement. Define solutions to meet performance, capability, and scalability needs.
Must have skills : Python (Programming Language)
Good to have skills : CommerceTools Commerce Platform
Minimum 3 year(s) of experience is required
Educational Qualification : 15 years full time education
Summary:
Design, build, and ship GenAI features end to end across web/backend stacks. You ll implement LLM powered experiences (chat, copilots, content automation), robust RAG pipelines, and API integrations using Python and modern front end frameworks, deploying on AWS/Azure/GCP.
Roles & Responsibilities:
- Implement GenAI features (chatbot flows, agent tools, content generation, summarization) with Python backends (FastAPI/Flask/Django) and React/Angular front ends.
- Build RAG pipelines: ingestion, chunking, embeddings, vector search, retrieval orchestration, response templating.
- Integrate cloud AI services: AWS Bedrock (Claude, Titan), Azure OpenAI (GPT 4o family), Google Vertex AI (Gemini) plus model endpoints from Hugging Face.
- Develop data connectors to S3/Blob/GCS, Kendra/Azure AI Search/Vertex Search, relational/NoSQL stores.
- Engineer prompt templates, tool use, guardrails, and evaluation harnesses (toxicity, hallucinations, latency, quality).
- Implement observability & MLOps hooks (OpenTelemetry, logging, tracing, CI/CD), model/config versioning, blue/green deployments.
- Write secure, testable code (unit/integration tests), perform code reviews, and contribute to reusable libraries/components.
Professional & Technical Skills:
- Strong Python (async, typing), REST/GraphQL APIs, microservices JavaScript/TypeScript for UI.
- LLMs & GenAI fundamentals: prompting, function/tool calling, structured outputs, evaluation.
- RAG: embeddings (Titan, text embedding ada/EP), vector DBs (Kendra/AI Search/OpenSearch/FAISS), retrieval strategies (hybrid).
- Cloud fluency:
- AWS: Bedrock, Lambda, S3, API Gateway, Step Functions, DynamoDB, Kendra, OpenSearch.
- Azure: OpenAI, Functions, Key Vault, Cosmos DB, Azure AI Search, App Service, AKS.
- GCP: Vertex AI, Cloud Run, Cloud Functions, BigQuery, Firestore.
- CI/CD (GitHub Actions/Azure DevOps), containers (Docker, Kubernetes), IaC (Terraform/CloudFormation/Bicep).
- Security: secret management (Key Vault/Secrets Manager), data privacy, prompt injection defenses.
- Experience with multi agent frameworks (LangGraph/LangChain), streaming (Server Sent Events/WebSockets).
- Front end design systems, accessibility, and performance tuning.
- Exposure to analytics/telemetry pipelines for model quality monitoring
Additional Information:
- Bachelor s/Master s in CS/Engineering (or equivalent).
- Typically 7–10 years total experience 2–4 years in applied GenAI/LLM solutions.
Project Role Description : Design and deliver technology architecture for a platform, product, or engagement. Define solutions to meet performance, capability, and scalability needs.
Must have skills : Python (Programming Language)
Good to have skills : CommerceTools Commerce Platform
Minimum 3 year(s) of experience is required
Educational Qualification : 15 years full time education
Summary:
Design, build, and ship GenAI features end to end across web/backend stacks. You ll implement LLM powered experiences (chat, copilots, content automation), robust RAG pipelines, and API integrations using Python and modern front end frameworks, deploying on AWS/Azure/GCP.
Roles & Responsibilities:
- Implement GenAI features (chatbot flows, agent tools, content generation, summarization) with Python backends (FastAPI/Flask/Django) and React/Angular front ends.
- Build RAG pipelines: ingestion, chunking, embeddings, vector search, retrieval orchestration, response templating.
- Integrate cloud AI services: AWS Bedrock (Claude, Titan), Azure OpenAI (GPT 4o family), Google Vertex AI (Gemini) plus model endpoints from Hugging Face.
- Develop data connectors to S3/Blob/GCS, Kendra/Azure AI Search/Vertex Search, relational/NoSQL stores.
- Engineer prompt templates, tool use, guardrails, and evaluation harnesses (toxicity, hallucinations, latency, quality).
- Implement observability & MLOps hooks (OpenTelemetry, logging, tracing, CI/CD), model/config versioning, blue/green deployments.
- Write secure, testable code (unit/integration tests), perform code reviews, and contribute to reusable libraries/components.
Professional & Technical Skills:
- Strong Python (async, typing), REST/GraphQL APIs, microservices JavaScript/TypeScript for UI.
- LLMs & GenAI fundamentals: prompting, function/tool calling, structured outputs, evaluation.
- RAG: embeddings (Titan, text embedding ada/EP), vector DBs (Kendra/AI Search/OpenSearch/FAISS), retrieval strategies (hybrid).
- Cloud fluency:
- AWS: Bedrock, Lambda, S3, API Gateway, Step Functions, DynamoDB, Kendra, OpenSearch.
- Azure: OpenAI, Functions, Key Vault, Cosmos DB, Azure AI Search, App Service, AKS.
- GCP: Vertex AI, Cloud Run, Cloud Functions, BigQuery, Firestore.
- CI/CD (GitHub Actions/Azure DevOps), containers (Docker, Kubernetes), IaC (Terraform/CloudFormation/Bicep).
- Security: secret management (Key Vault/Secrets Manager), data privacy, prompt injection defenses.
- Experience with multi agent frameworks (LangGraph/LangChain), streaming (Server Sent Events/WebSockets).
- Front end design systems, accessibility, and performance tuning.
- Exposure to analytics/telemetry pipelines for model quality monitoring
Additional Information:
- Bachelor s/Master s in CS/Engineering (or equivalent).
- Typically 7–10 years total experience 2–4 years in applied GenAI/LLM solutions.
职位要求
15 years full time education
