AI Engineer

InfraCents Pvt. Ltd.; India; Remote; INR 1,100,000 - 1,700,000; Full time

Job summary

InfraCents Pvt. Ltd. seeks an AI Engineer to design and ship LLM-powered products for clients, from retrieval-augmented assistants to agents that automate workflows.

You will own features end-to-end with emphasis on evaluation, safety and cost, treating AI as software engineering with solid APIs, observability and fail-safe design.

Qualifications

• 1–3 years in software or ML engineering, including LLM applications in production or pilots.
• Strong Python and API development skills, not just notebooks.
• Evaluation-first mindset with data-driven quality checks.
• Understanding of Responsible AI: privacy, security and risk awareness.
• Clear communication to non-technical stakeholders about AI capabilities and limits.

Responsibilities

• Build RAG pipelines, AI assistants and tool-using agents for client use cases.
• Design prompts, structured outputs and robust tool schemas.
• Integrate AI features into backend services via clean APIs.

Skills

Python

API development

LLM applications in production

Communication

Education

Bachelor's or Master's in Computer Science or Engineering

Tools

FastAPI

PostgreSQL

Redis

Docker

Node.js/TypeScript

Job description

About InfraCents

InfraCents is a software engineering company. We design and build software products, run the cloud platforms they live on, build AI systems and data pipelines, and keep all of it secure, reliable and cost-efficient in production. Our engineers work directly with clients, own outcomes end to end and are trusted with real production systems. We value clear thinking, honest communication and work that still holds up a year after it ships.

Role Overview

The AI Engineer will design and ship LLM-powered products for clients, from retrieval-augmented assistants and document intelligence to agents that use tools and automate workflows. You will own features from prototype to production, with a strong focus on evaluation, safety and cost.

We are looking for an engineer who treats AI as software engineering: measurable quality, solid APIs, observability and failure handling, not just impressive demos.

Key Responsibilities: LLM Application Development

• Build RAG pipelines, AI assistants and tool-using agents for client use cases.
• Design prompts, structured outputs and tool schemas that are robust and testable.
• Integrate AI features into backend services and products through clean APIs.

Key Responsibilities: Retrieval & Data

• Build ingestion pipelines: parsing, chunking, embedding, metadata and access control.
• Implement hybrid search and reranking, and tune retrieval quality.
• Keep knowledge sources fresh, versioned and traceable to their origin.

Key Responsibilities: Evaluation & Quality

• Create evaluation datasets and automated evals for accuracy, groundedness and safety.
• Track quality regressions when prompts, models or data change.
• Implement guardrails: input/output validation, PII redaction and prompt-injection defences.

Key Responsibilities: Production & Operations

• Monitor latency, cost and quality in production with tracing and dashboards.
• Optimise model choice, caching and batching to control cost.
• Explain AI capabilities, limits and trade-offs clearly to clients.

Who We're Looking For

• Experience: 1-3 years in software or ML engineering, including LLM applications in production or serious pilots.
• Engineering depth: Strong Python and API development skills, not just notebooks.
• Evaluation-first: Measures quality with data before declaring something works.
• Responsible AI: Understands hallucination, data privacy and security risks of LLM systems.
• Communication: Can explain what AI can and cannot do to non-technical stakeholders.

Technical Expertise Required

• Models: Anthropic Claude, OpenAI GPT, Google Gemini, Open-weight models (Llama, Mistral, Qwen)
• Cloud AI Platforms: AWS Bedrock, Azure OpenAI, Google Vertex AI
• Orchestration & Agents: LangGraph, LlamaIndex, Claude Agent SDK / OpenAI Agents SDK, Model Context Protocol (MCP)
• Retrieval: pgvector, Pinecone, Qdrant / Weaviate, OpenSearch / Elasticsearch, Rerankers
• Evaluation & Observability: Langfuse, LangSmith, Ragas, promptfoo / DeepEval
• Engineering: Python, FastAPI, TypeScript / Node.js, PostgreSQL, Redis, Docker

Nice to Have

• Experience with document AI: OCR, layout parsing and structured extraction.
• Experience with voice or multimodal (vision) models.
• Exposure to fine-tuning or serving open-weight models.
• Experience deploying AI services on Kubernetes or serverless platforms.

Qualifications

• Experience: 1-3 years in software, ML or AI engineering.
• Education: Bachelor's or Master's degree in Computer Science, Engineering or a related field, or equivalent practical experience.