20 Nov
|
VeriSmart.ai
|
Gurugram
20 Nov
VeriSmart.ai
Gurugram
Key Responsibilities:- Design, build and maintain high-performance REST APIs and microservices using FastAPI, Pydantic and async Python.- Own the scalability of backend services for high-traffic web and mobile products, including capacity planning, load testing, caching and horizontal scaling.- Keep latency-critical APIs fast under peak load.- Develop NLP and LLM-powered features such as text classification, extraction, summarization, semantic search and conversational assistants.- Build LLM applications and agentic workflows with LangChain and LangGraph, including tool use, state management and multi-step orchestration.- Implement retrieval-augmented generation (RAG) pipelines covering chunking, embeddings, vector search and prompt design.- Evaluate and improve model quality, latency and cost through prompt iteration, evaluation datasets and monitoring.- Integrate third-party APIs and messaging channels (email, SMS, WhatsApp, push notifications and OTP) reliably, with timeouts, retries and background job queues.- Write clean, tested and well-documented code, including unit and integration tests, and participate in code reviews.- Containerize and deploy services, and work with DevOps on CI/CD, observability and production support.- Collaborate with product, data and frontend teams to turn requirements into reliable features.Required Qualifications:- 3 - 4 years of professional experience in Python software development.- Proven experience scaling a live product: backend systems serving 50,000 to 1 million website users and supporting mobile apps with more than 1 million downloads, including traffic growth and peak-load events.- Robust understanding of backend performance: async I/O and event-loop behavior, caching with Redis, database indexing and query optimization, connection pooling, and running multi-worker services (Gunicorn/Uvicorn)
behind a load balancer.- Hands-on experience building production APIs with FastAPI, including async programming, dependency injection and Pydantic validation.- Working knowledge of machine learning, with a focus on NLP: embeddings, transformers, text classification and information extraction.- Practical experience with LLMs: prompt engineering, structured outputs, function calling, and handling hallucinations, latency and cost.- Experience with LangChain and LangGraph for building LLM applications and agent workflows.- Hands-on experience with MongoDB at scale, including schema design, indexing, aggregation pipelines and query performance tuning.- Solid understanding of relational databases (PostgreSQL, MySQL or Oracle) and SQL.- Hands-on experience with Docker for containerizing, running and deploying services.- Experience building secure systems that handle personal data: token-based authentication (JWT, OAuth2), role-based access control, encryption, audit logging and awareness of data-protection regulations such as India's DPDP Act.- Familiarity with Git, automated testing (pytest), CI/CD pipelines (Jenkins, GitHub Actions or similar) and code review practices.- Good understanding of REST API design, authentication and software engineering fundamentals.- Strong communication skills and the ability to work independently in a team.Good to Have:- Experience building backend services in Java or Node.js.- Experience with vector databases such as pgvector, Pinecone, Qdrant or Chroma, and RAG evaluation.- Experience integrating hardware or IoT devices, including offline or on-premise deployments.- Exposure to LLM evaluation and observability tools such as LangSmith.- Experience fine-tuning or serving open-source models (Hugging Face, PyTorch).- Knowledge of MLOps practices, CI/CD and monitoring.- Contributions to open-source projects or a public portfolio. (ref:hirist.tech)
📌 Verismart.ai - Backend Engineer (Gurugram)
🏢 VeriSmart.ai
📍 Gurugram