Technical Architect - AI
Posted Updated
Description
Job Description :
Role & Responsibilities :
- Architectural Leadership : You are the go-to person for all AI/ML infrastructure, Agentic workflows, and RAG technology implementations at Lendingkart.
- Strategic Decisions : Play a crucial role in driving AI product and technology decisions, from model selection (SLMs vs. LLMs) to operational deployments.
- Roadmap & Hiring : Show high levels of ownership in driving the AI roadmap, scalable architectures, and hiring top-tier AI and Data Engineers.
- Code & Design Quality : Actively review code, lead design reviews, and guide architectural discussions for multi-agent systems, prompt engineering frameworks, and MCP server integrations.
- Best Practice Adoption : Drive best practices around Generative AI, fine-tuning methodologies (LoRA/QLoRA), model evaluation (Evals), guardrails, and latency optimization.
- Innovation & Experimentation : Experiment with novel AI frameworks, open-source models, and toolings to continuously drive business efficiency and customer impact.
- Tech Ambassador : Represent Lendingkart in external technical forums and AI conferences as a key technology ambassador.
Preferred candidate profile :
Core AI & Generative AI Expertise (Hands-On) :
- LLMs & SLMs : Practical experience in fine-tuning, serving, and evaluating open-source (Llama, Mistral, Qwen) and proprietary models using LoRA, QLoRA, PEFT, and quantization techniques.
- Agentic AI & Frameworks : Production experience building autonomous, multi-agent frameworks using tools like LangChain, LangGraph, LlamaIndex, AutoGen, or CrewAI.
- MCP Server Integrations : Expertise in integrating Model Context Protocol (MCP) servers/tools to securely expose enterprise context and tools to LLMs.
- RAG Pipelines & Vector DBs : Proven track record of architecting scalable enterprise RAG systems utilizing vector databases (Pinecone, Milvus, Qdrant, PGVector) with advanced retrieval techniques (Hybrid Search, GraphRAG, Query Rewriting, Re-ranking).
- Prompt Engineering & Evals : Mastery of advanced prompt strategies (Chain-of-Thought, ReAct, Few-Shot) and evaluation/observability stacks (Ragas, TruLens, DeepEval, LangSmith, Langfuse).
Engineering, Cloud & Security :
- Languages & Frameworks : Highly proficient in Python (PyTorch, Hugging Face, vLLM, Ollama, FastAPI) and backend systems in Golang, Java, or JavaScript.
- Cloud & Cloud-Native : Experience with AWS (Bedrock, SageMaker) and/or GCP (Vertex AI), alongside containerization (Docker, Kubernetes).
- AI Security & Guardrails : Hands-on knowledge of AI guardrails (NeMo Guardrails, Llama Guard, OWASP LLM Top 10), data privacy, PII masking, and red-teaming.
- DevOps & Infrastructure : Experience with CI/CD, Infrastructure as Code (Terraform), and LLMOps monitoring/alerting.
Skills
- Agentic AI
- AI
- AutoGen
- AWS
- CI/CD
- Cloud
- Cloud Native
- Containerization
- CrewAI
- DevOps
- Docker
- FastAPI
- Fine Tuning
- GCP
- Generative AI
- Go
- Hugging Face
- Infrastructure as Code
- Java
- JavaScript
- Kubernetes
- LangChain
- Langfuse
- LangGraph
- LangSmith
- LlamaIndex
- LLM
- LLMOps
- Machine Learning
- MCP
- Milvus
- Mistral
- Model Evaluation
- Observability
- Ollama
- OWASP
- PEFT
- pgvector
- Pinecone
- Prompt Engineering
- Python
- PyTorch
- Qdrant
- Quantization
- SageMaker
- Terraform
- Vector Databases
- Vertex AI
- vLLM
Sign in to read and join the discussion.