Senior AI Engineer (Full-Stack) (Remote)

India·PuneRemotesenior

Machine Learning EngineerData

0 views0 saves0 applied

Apply Now

Quick Summary

Key Responsibilities

Token usage logging Cost tracking Fallback detection Performance monitoring Regression test suites Build evaluation pipelines and golden test suites to ensure AI quality and consistency.

Requirements Summary

Pinecon

Technical Tools

Machine Learning EngineerData

Employment Type: Full-Time
Location: Remote

About the Role

~1 min read

We are looking for a Senior AI Engineer with strong full-stack capabilities to join our engineering team. You will work across multiple client-facing projects that sit at the intersection of AI, data, and web—starting with a live production AI agent platform.

You should be comfortable owning features end-to-end, from LLM pipeline design through to a polished frontend experience.

Responsibilities

~1 min read

Design and build AI agent pipelines, including multi-node LangGraph graphs.
Implement intent routing, multi-turn conversational context, session state management, and tool integrations.
Develop multi-step reasoning pipelines and graph-based agent workflows.

LangGraph or equivalent graph-based agent frameworks.
Multi-step reasoning pipelines.
Tool usage and orchestration.
State management and conversational workflows.

Build and maintain Retrieval-Augmented Generation (RAG) systems.
Design vector search architectures, embedding pipelines, retrieval grounding, and chunking strategies.
Implement hallucination mitigation techniques and retrieval evaluation frameworks.

Integrate and optimize Large Language Models (LLMs) including OpenAI, Gemini, and Anthropic.
Develop structured output workflows using JSON schemas.
Create effective prompt engineering strategies, few-shot examples, and context window management solutions.
Build provider-neutral client architectures to support multiple LLM vendors.

Design, develop, and deploy end-to-end product features.
Build scalable FastAPI backends and React/Next.js frontends.
Implement Server-Sent Events (SSE) streaming and REST API contracts.
Deliver production-ready UI features independently without requiring dedicated frontend support.

React
Next.js
TypeScript
Modern frontend architecture
API integration and state management

Own LLM observability including:
- Token usage logging
- Cost tracking
- Fallback detection
- Performance monitoring
- Regression test suites
Build evaluation pipelines and golden test suites to ensure AI quality and consistency.

Collaborate directly with clients and stakeholders to understand business requirements.
Translate requirements into scalable, maintainable software solutions.
Keep technical documentation, specifications, and test coverage aligned with product changes.

End-to-end RAG pipeline design and implementation.
Experience with vector databases such as:
- Pinecone
- Qdrant
- pgvector
- Weaviate
Chunking strategies and retrieval optimization.
Retrieval evaluation methodologies.

OpenAI, Gemini, and Anthropic SDKs.
Prompt engineering and prompt optimization.
Structured JSON outputs.
Context window management.
Multi-provider LLM integrations.

Python 3.12
FastAPI
Async Python
Pydantic
SQLite
PostgreSQL
Redis
Pytest

Evaluation pipelines
Golden datasets and test suites
Regression tracking
Model performance monitoring

ArcGIS REST APIs
GeoJSON
MapLibre GL JS
Spatial queries

(Strong advantage for initial project assignments.)

Recharts
D3.js
Equivalent charting libraries

Docker
Azure
AWS
CI/CD pipelines
OIDC Authentication

Ability to understand and interpret Figma designs.
Evaluate trade-offs between engineering effort and business value.
Deliver solutions aligned with business objectives.

Layer	Technology Stack
Agent Frameworks	LangGraph, LangChain
LLM Providers	Gemini, OpenAI, Anthropic
Backend	Python 3.12, FastAPI, SQLite, Redis
Frontend	Next.js 15, React 19, TypeScript, Zustand
Data & Visualization	Recharts, GeoJSON, MapLibre GL JS
Infrastructure	Docker, Azure Pipelines, Azure AD

Someone who can independently own a feature from requirements gathering to production deployment.
Strong full-stack engineering capabilities with no hand-holding required between backend and frontend development.
Strong engineering judgment and the ability to push back when shortcuts introduce hallucination risks, reliability issues, or technical debt.
Comfortable working in ambiguous environments with evolving client requirements.
Experience delivering software in real-world production environments.
Excellent communication skills with the ability to explain AI system behavior and limitations to non-technical stakeholders.
Strong documentation and testing discipline.