AI Engineer

SANSKRITI; New Jersey; On-site; USD 150,000 - 210,000; Full time

Job summary

Sanskriti is building a new generation of AI-powered digital experiences. We seek an AI Engineer to design production-grade AI/ML systems, orchestrate long-running workflows and integrate multiple models for reliable operations.

You will own evaluation, provisioning of robust pipelines and observability in a fast-paced environment. Join a team that blends AI, software engineering and production systems to transform experimental ideas into scalable, secure solutions.

Qualifications

• Bachelor’s or Master’s in CS/Engineering or related field.
• 5–8 years in software/ML/AI engineering.
• Strong Python programming fundamentals.
• Hands-on experience with LLMs or generative models.
• Experience designing multi-step AI workflows.
• Familiarity with cloud infrastructure and production services.

Responsibilities

• Design and build production-grade AI/ML systems using LLMs.
• Create AI-powered workflows and agentic systems for reliable task execution.
• Develop orchestration pipelines for long-running, asynchronous workloads.
• Integrate multiple AI models and providers via clean interfaces.
• Build model selection, routing, evaluation and fallback mechanisms.

Skills

Python

AI/ML engineering

Distributed systems

Asynchronous workflows

Observability

Prompt engineering

Education

Bachelor’s/Master’s degree in CS/Engineering/AI

Tools

Temporal

Airflow

LangChain

Docker

AWS

PostgreSQL

OpenTelemetry

MLflow

Job description

Sanskriti is building a new generation of digital experiences around Indian culture, stories and knowledge. Our goal is to make this rich cultural heritage more accessible, engaging and relevant to audiences in India and around the world.

We are combining technology, creativity and AI to build new ways for people to discover, experience and engage with Indian culture at scale.

We are at an early stage of building the technology and systems that will power this vision, and we're looking for people who want to build from first principles rather than simply plug existing AI tools together.

About the Role

We are looking for an AI Engineer to help build and scale an AI-powered production system from the ground up.

This is a hands-on engineering role for someone who enjoys working at the intersection of AI, software engineering, automation and production systems. You will work with modern AI models and agents, build reliable workflows around them, and take AI capabilities from experimentation to production.

You will have significant ownership over how AI systems are designed, evaluated, integrated and operated.

What You'll Do

• Design and build production-grade AI/ML systems using LLMs and generative AI models.
• Build AI-powered workflows and agentic systems that can execute structured tasks reliably.
• Develop orchestration and execution pipelines for long-running, asynchronous and multi-step workloads.
• Integrate multiple AI models and external providers through clean, replaceable interfaces.
• Build systems for model selection, routing, evaluation and fallback based on quality, cost, latency and operational constraints.
• Develop structured prompting, compilation and context-management systems rather than relying only on free-form prompts.
• Build automated evaluation, quality-control and validation pipelines for AI-generated outputs.
• Design robust retry, failure-handling, escalation and recovery mechanisms for AI workflows.
• Build strong observability and telemetry around models, agents, workflows and production outcomes.
• Maintain detailed experiment, model, prompt and output lineage to make AI systems reproducible and debuggable.
• Work with structured data, APIs, databases, object storage and asynchronous job systems.
• Build and maintain evaluation datasets, benchmarks and feedback loops to continuously improve system performance.
• Ensure AI systems are reliable, secure, cost-efficient and capable of operating at production scale.
• Work closely with product and engineering teams to translate ambiguous problems into working AI systems.

What We're Looking For

• Education: Bachelor’s/Master’s degree in Computer Science, Engineering, AI/ML, or a related field from a Tier 1 institution.
• 5-8 years of experience in software engineering, ML engineering, AI engineering or a closely related field
• Strong programming skills in Python.
• Strong software engineering fundamentals: APIs, databases, testing, version control, system design and debugging.
• Hands‑on experience building applications with LLMs or generative AI models.
• Good understanding of agents, tool calling, structured outputs, RAG, embeddings and prompt engineering.
• Experience designing multi-step AI workflows rather than only making individual API calls.
• Understanding of asynchronous processing, queues, workers and distributed/long-running workflows.
• Experience integrating and working with multiple AI models or model providers.
• Strong understanding of evaluation and the challenges of measuring AI quality in production.
• Familiarity with cloud infrastructure and deploying production services.
• Comfortable working with incomplete requirements, experimenting quickly and turning prototypes into reliable systems.

Good to Have

• Experience with workflow orchestration frameworks such as Temporal, Airflow or similar systems.
• Experience with agent frameworks such as Pydantic AI, LangGraph, LangChain or similar.
• Experience with model-serving or generative-AI infrastructure.
• Experience with MLflow, OpenTelemetry or other experiment/observability systems.
• Experience with Docker and cloud platforms such as AWS/GCP/Azure.
• Experience with PostgreSQL and object storage such as S3.
• Experience building evaluation frameworks, benchmarks or automated QC systems.
• Experience working with multimodal AI — text, image, audio or video.
• Familiarity with model routing, inference optimisation, GPU workloads or cost optimisation.
• Experience building systems where reproducibility, provenance, security and auditability matter.

What We Value

We are looking for engineers who:

• Build, rather than just experiment.
• Understand that an AI demo and a production AI system are very different things.
• Can reason about quality, cost, latency and reliability together.
• Prefer measurable evaluation over subjective claims about model performance.
• Design systems so that models and vendors can be replaced without rebuilding everything.
• Care about clean interfaces, testing, observability and failure handling.
• Can balance rapid experimentation with production engineering discipline.

Why This Role

You will be part of a team building an AI-native production system from the ground up, with the opportunity to work across agents, generative models, orchestration, evaluation, infrastructure and intelligent automation.

This is not a role where you'll simply be writing prompts or connecting APIs. You'll be building the engineering system around AI that makes it reliable and usable in production.