Artificial Intelligence Engineer

Onward Technologies; Bengaluru; On-site; INR 2,500,000 - 4,000,000; Full time

Job summary

Onward Technologies is seeking an AI Engineer to build AI-assisted Systems Engineering solutions that convert legacy locomotive information into structured SFS, SDS and traceability artifacts. You will leverage LLMs, RAG, semantic search and controlled AI workflows to deliver end-to-end pipelines.

You will implement human-in-the-loop review, confidence scoring and grounding, while integrating with ALM/PLM systems and engineering tools. A background in Rail/Automotive domains is desirable.

Qualifications

• 4–8 years of experience in AI/ML engineering or related field.
• Experience with Generative AI, LLMs and RAG workflows.
• Hands-on Python development and API design experience.
• Familiarity with embedding-based retrieval and vector databases.

Responsibilities

• Develop AI pipelines to extract and structure engineering data from legacy sources (PDF/DOCX/XLSX/XML).
• Create RAG-based workflows for retrieving engineering evidence.
• Implement requirement extraction, classification and transformation workflows.
• Generate SFS/SDS candidates using structured outputs.
• Implement semantic search and vector-based retrieval.
• Develop traceability, engineering knowledge workflows and grounding.
• Incorporate human-in-the-loop review and approval mechanisms.
• Build evaluation datasets and benchmark AI outputs.

Skills

Python

Generative AI

LLMs

RAG

LangChain

LangGraph

LlamaIndex

embeddings

semantic search

vector databases

PostgreSQL/pgvector

FAISS

Pydantic

PyMuPDF

python-docx

openpyxl

Neo4j

FastAPI

REST APIs

Education

Bachelor's or Master's in CS/AI/Engineering

Tools

PostgreSQL/pgvector

FAISS

Neo4j

FastAPI

pyMuPDF

python-docx

openpyxl

Job description

Experience: 4 to 8 years

Domain: Generative AI / NLP / Engineering AI / Digital Engineering

Role Overview

We are looking for an AI Engineer to develop an AI-assisted Systems Engineering solution that transforms legacy locomotive engineering information into structured SFS, SDS and traceability artifacts, using LLMs, RAG, semantic search, engineering knowledge and controlled AI workflows.

Key Responsibilities

• Develop AI pipelines for extracting and structuring engineering information from PDF, DOCX, XLSX, XML and other legacy sources.
• Develop RAG-based workflows for retrieving relevant engineering evidence.
• Implement requirement extraction, classification and transformation workflows.
• Develop AI-assisted generation of SFS/SDS candidates using structured outputs.
• Implement semantic search, embeddings and vector-based retrieval.
• Develop traceability recommendation and engineering knowledge workflows.
• Integrate engineering rules and deterministic validation into AI workflows.
• Implement human-in-the-loop review and approval mechanisms.
• Develop confidence scoring, source grounding and hallucination detection/mitigation approaches.
• Build evaluation datasets and benchmark AI outputs against the manual Systems Engineering baseline.
• Develop APIs and automation using Python.
• Support integration with ALM/PLM systems and engineering tools.

Required AI / Technical Skills

• Strong hands-on Python development.
• Practical experience with Generative AI, LLMs and RAG.
• Experience with frameworks such as LangChain, LangGraph, LlamaIndex or equivalent.
• Experience with embeddings, semantic search and vector databases.
• Knowledge of PostgreSQL/pgvector, FAISS or similar.
• Experience with structured LLM outputs and Pydantic.
• Experience with document processing libraries such as PyMuPDF, Docling, python-docx, openpyxl or equivalent.
• Understanding of knowledge graphs; Neo4j exposure preferred.
• Experience developing REST APIs using FastAPI or similar.
• Understanding of AI evaluation, grounding, hallucination, confidence and human-in-the-loop workflows.
• Exposure to Systems Engineering, Requirements Engineering or engineering-domain AI is highly desirable.

Rail / Engineering Knowledge

• Exposure to Rail, Automotive, Aerospace, Industrial or other safety-critical engineering domains preferred.
• Understanding of requirements, functional architecture, traceability and V&V is an advantage.

Education: Bachelor's/Master's degree in Computer Science, AI/ML, Electronics, Electrical, Engineering or related discipline.