Data Engineer

Recrew AI; Bengaluru Urban; On-site; INR 1,500,000 - 2,500,000; Full time

Benefits offered by this job

Early ownership

Seed-stage startup

Impact from day one

Job summary

Recrew AI in Bengaluru is seeking an AI Data Engineer to own the full data lifecycle—from raw sources to model-ready inputs. You will design, ship, and operate batch and streaming pipelines that power AI products and agents, ensuring reliability at scale.

You will extract structured data from diverse sources (PDFs, documents, web pages, APIs) and enforce data quality with robust validation and schema checks. Collaboration with ML and backend teams is essential for production-ready systems.

Qualifications

• B.Tech / B.E. / M.Tech in CS or related field from IITs/NITs/BITS.
• 2–3 years of hands-on data pipeline experience in production.
• Strong Python and SQL skills for production data systems.
• Hands-on with LLM APIs for extraction/classification/transformations.
• Experience with AI agents and agent-driven workflows.
• Proven ownership of data quality in live systems.

Responsibilities

• Design, ship, and maintain batch and streaming pipelines that feed training, evaluation, RAG, and agent workflows end-to-end.
• Extract structured data from PDFs, scanned documents, web pages, APIs, and internal systems using LLM-based and traditional extraction methods.
• Define and enforce data quality: validation rules, schema checks, deduplication, anomaly detection, and extraction evals before data reaches users or models.
• Build tools, data interfaces, and guardrails for AI agents; process and audit agent outputs; leverage coding agents to accelerate pipelines.
• Own production operations for pipelines: monitoring, alerting, retries, cost and latency tracking, and on-call response.
• Manage embedding pipelines: chunking strategies, embedding generation, vector store indexing, and refresh cycles.
• Collaborate with ML, product, and backend engineers to translate data requirements into reliable, documented systems.

Skills

Python

SQL

LLM APIs

AI agents

Data quality

Data pipelines

Education

CS degree (IIT/NIT/BITS)

Tools

Kafka

Kinesis

dbt

Great Expectations

Spark

DuckDB

Job description

Industry: Information Technology & Services / Artificial Intelligence / Critical Infrastructure

About Company

The company is a research-first AI startup incubated at the Indian Institute of Science (IISc). It builds advanced AI for the planning and operations of critical networks.

Its core technology is a World Model that augments LLMs to reason about network state and dependencies, even with limited or noisy data. The founding team has raised over $100M across earlier ventures and published at leading AI research venues.

The company has just closed its seed round and is assembling a lean, ambitious team.

Position Overview

As AI Data Engineer, you will own the full data lifecycle — from raw, messy sources to model-ready inputs — that powers the company's AI products and agents. You will design, ship, and operate batch and streaming pipelines in production, not just prototypes. Working at the intersection of data engineering and applied AI, you will build the infrastructure that makes the company's models and agents reliable at scale.

Role & Responsibilities

• Design, ship, and maintain batch and streaming pipelines that feed training, evaluation, RAG, and agent workflows end-to-end.
• Extract structured data from PDFs, scanned documents, web pages, APIs, and internal systems using LLM-based and classical extraction methods.
• Define and enforce data quality: validation rules, schema checks, deduplication, anomaly detection, and extraction evals before data reaches users or models.
• Build tools, data interfaces, and guardrails for AI agents; process and audit agent outputs; use coding agents to accelerate your own pipeline work.
• Own production operations for pipelines you build — monitoring, alerting, retries, cost and latency tracking, and on-call response.
• Manage embedding pipelines: chunking strategies, embedding generation, vector store indexing, and refresh cycles.
• Collaborate with ML, product, and backend engineers to translate data requirements into reliable, documented systems.

Must Have Criteria

• B.Tech / B.E. / M.Tech in Computer Science or a related field from an IIT or NIT/BITs.
• 2–3 years of hands-on experience building and shipping data pipelines to production — not just notebooks or prototypes.
• Strong Python and SQL, applied to real production data systems.
• Hands-on experience with LLM APIs for extraction, classification, or transformation, including prompt design and structured outputs.
• Experience building or working with AI agents — tool calling, agent frameworks, or agent-driven workflows.
• Demonstrated ownership of data quality in live systems: validation, testing, and debugging bad data.

Nice to Have

• Experience with vector databases and RAG pipelines.
• Document AI or OCR work — layout parsing, table extraction, or multimodal models.
• Building LLM evaluation sets or human-in-the-loop review workflows.
• Streaming systems such as Kafka or Kinesis.
• Data tooling such as dbt, Great Expectations, Spark, or DuckDB.
• Prior experience at an early-stage or fast-moving startup.

What We Offer

• Early ownership at a seed-stage, research-first AI company incubated at IISc — with real influence over technical direction.
• Work on frontier AI problems at the intersection of LLMs, agents, and critical network infrastructure.
• Collaborate directly with a founding team that has raised $100M+ and published at leading AI venues.
• Lean, high-trust environment where your work ships to production and has measurable impact from day one.