AI Platform Engineer / ML Infrastructure Engineer

US / India Deep-Tech Startup | Chennai & San Jose

(Full-time

• On-site / Hybrid)

About Us

We are a US / India deep-tech startup based in San Jose, CA and Chennai , building next-generation AI-driven medical intelligence systems .

Our work sits at the intersection of :

real-time data systems,

safety-critical engineering,

applied machine learning, and

modern information retrieval (vector search, graph reasoning, RAG).

We operate in a domain where reliability, interpretability, and intelligent automation truly matter.

If you enjoy working on technically challenging problems that have real-world impact, you’ll feel at home here.

Role Overview

We are looking for an engineer who can design end-to-end AI and data platforms —from data ingestion to inference—across vector search, graph-based modeling, ML infrastructure, and cloud-native deployment.

This role blends modern ML Ops, data engineering, knowledge retrieval, and systems design.

If you are comfortable thinking across multiple layers of the stack and enjoy solving open-ended problems, we’d love to speak with you.

What You’ll Work OnData & Retrieval Systems

Build and optimize pipelines using vector search , embeddings, and metadata-aware retrieval.

Experiment with graph / knowledge-based systems (e.g., RDF stores, property graphs, knowledge graphs).

Help design hybrid retrieval workflows (vector + graph + rules).

ML & Inference Infrastructure

Set up and manage ML pipelines , model serving endpoints, and inference layers.

Work on streaming / batch data flows across AWS / GCP.

Build observability and monitoring for data quality, drift, and safety.

Cloud & Infra Automation

Deploy cloud resources using AWS + Terraform (EC2, S3, RDS / Neptune, networking, IAM).

Manage containerized services (Docker + optional Kubernetes experience).

Ensure scalable, secure, and reproducible infrastructure.

Systems-Level Problem Solving

Work cross-discipline to design architectures that integrate real-time data, ML models, graph semantics, and domain constraints.

Bring strong reasoning on system trade-offs, reliability, safety, and performance.

What We’re Looking For

We don’t expect you to know everything — but you should be strong in some of these areas and curious about the rest :

Core Skills

Experience with vector databases (pgvector, Pinecone, Weaviate, Milvus) or strong understanding of semantic search.

Experience with graph databases or knowledge representation (Neo4j, Neptune, RDF, SPARQL, property graphs).

Solid grounding in Python , data pipelines, and model-serving frameworks.

Hands-on experience with AWS services (EC2, S3, IAM, networking).

Practical understanding of Terraform or infra-as-code.

Bonus Skills (not required but appreciated)

Experience building or maintaining RAG pipelines.

Understanding of ML observability and safety.

Exposure to healthcare or high-reliability systems.

Interest in temporal reasoning, anomaly detection, or multimodal data.

Who You Are

You think like a systems engineer—holistic, curious, rigorous.

You enjoy technical depth more than buzzwords.

You are comfortable exploring ambiguous problems.

You want to work on ML infrastructure that actually matters in the real world.

You appreciate startups where engineers have autonomy and ownership.

Location

Chennai preferred, hybrid options available.

Qualifications & Experience Required Qualifications We are looking for engineers with strong foundational experience across data systems, ML infrastructure, or cloud-native platforms. You should have :

Solid hands-on experience with Python and at least one ML / DS workflow.

Working knowledge of cloud services (AWS preferred; GCP / Azure okay).

Experience setting up or maintaining data pipelines (batch or streaming).

Exposure to vector search , embeddings, or semantic retrieval systems.

Good understanding of databases — SQL, NoSQL, or graph.

Experience deploying or managing infra using Terraform or other IaC frameworks.

Ability to design and reason about systems (latency, scaling, reliability, security).

Strong debugging and problem-solving mindset.

Preferred Experience (Good to Have, Not Mandatory) These are not hard requirements — but if you have them, we’d love to hear about it :

Experience with graph databases (Neo4j, Neptune, RDF, SPARQL, ArangoDB).

Familiarity with RAG pipelines , retrieval frameworks, or knowledge graphs.

Hands-on with Docker , containerized ML workloads, or microservices.

Exposure to ML Ops tools (SageMaker, Vertex, MLFlow, Kubeflow, BentoML).

Understanding of vector DBs like Pinecone, Weaviate, Milvus, or pgvector.

Knowledge of data modeling , schema design, or graph-based reasoning.

Experience building systems that require safety, accuracy, and auditability .

Interest in healthcare, IoT, or real-time monitoring systems.

Experience Range We expect candidates to have a meaningful track record building or deploying technical systems.

Typically this means 4–10 years of experience , depending on depth and exposure.

Why Join Us

You will work on deeply meaningful problems at the frontier of AI, safety, and intelligent automation.

This is not a typical “ML model training” role — it’s a platform engineering role for someone who likes to architect, design, and build real systems end-to-end.