01
TRCD-26-04117
Senior Data Scientist / AI Engineer (Computer Vision, OCR/Document AI, Multimodal LLM, RAG/Graph-RAG)
Hands-on Senior Data Scientist/AI Engineer to build and productionize AI for automated understanding and validation of complex 2D engineering drawings (PDF/DWG/DXF-derived). The role spans computer vision (layout/region detection, detection/segmentation, symbol recognition), OCR/Document AI, multimodal LLM workflows with structured JSON outputs and guardrails, and RAG/Graph-RAG over engineering standards using knowledge graphs (e.g., Neo4j). You will implement hybrid validation that combines deterministic rule engines with LLM reasoning, create explainable findings with citations, and take models into production as Python/FastAPI services with Docker/Kubernetes on NVIDIA GPU infrastructure, supported by MLOps practices (versioning, tracking, registry, monitoring).
Position summary
- Location
- India
- Workplace
- On-site
- Employment
- Contract
- Experience
- Minimum 7 years and Maximum 12 years
Role overview
Why This Role Matters.
Hands-on Senior Data Scientist/AI Engineer to build and productionize AI for automated understanding and validation of complex 2D engineering drawings (PDF/DWG/DXF-derived). The role spans computer vision (layout/region detection, detection/segmentation, symbol recognition), OCR/Document AI, multimodal LLM workflows with structured JSON outputs and guardrails, and RAG/Graph-RAG over engineering standards using knowledge graphs (e.g., Neo4j). You will implement hybrid validation that combines deterministic rule engines with LLM reasoning, create explainable findings with citations, and take models into production as Python/FastAPI services with Docker/Kubernetes on NVIDIA GPU infrastructure, supported by MLOps practices (versioning, tracking, registry, monitoring).
Your Impact
Deliver Enterprise Value
Help organisations solve complex business problems through modern technology, consulting expertise and measurable outcomes.
Collaboration
Work Across Teams
Collaborate with consultants, architects, engineers and client stakeholders throughout the project lifecycle.
Growth
Learn Continuously
Gain exposure to enterprise technologies, certifications, mentoring and real-world project experience.
Career Path
Grow With Ubique
Build a long-term consulting career with opportunities to take on greater responsibility and leadership over time.
Responsibilities
What You'll Be Doing.
Every role at Ubique contributes directly to solving meaningful business challenges for our clients.
02
Develop Computer Vision models for region detection, layout understanding, object detection, segmentation, symbol recognition, view classification and annotation localization.
03
Design OCR/document-processing pipelines for dense technical drawings using tools such as PaddleOCR, Docling, OpenCV, YOLO, Vision Transformers or equivalent frameworks.
04
Build multimodal/LLM workflows for engineering interpretation, structured extraction, contextual reasoning and JSON-based outputs with strong guardrails.
05
Develop RAG and Graph-RAG solutions over engineering standards, specifications, component rules and business-unit guidelines using vector search and Knowledge Graphs such as Neo4j.
06
Implement hybrid validation combining deterministic rule engines, symbolic checks and LLM reasoning; ensure exact geometric/tolerance checks are not delegated solely to LLMs.
07
Create explainable review findings including violated rule, source/citation, confidence, rationale and recommended corrective action.
08
Design Human-in-the-Loop feedback mechanisms to capture accept/reject/edit/override decisions and improve models, prompts, retrieval and rules.
09
Define accuracy benchmarks and golden datasets; track precision, recall, F1, IoU, OCR CER/WER, extraction accuracy, retrieval quality, validation accuracy and hallucination rate.
10
Perform failure analysis, prompt/model benchmarking, regression testing and continuous accuracy improvement.
11
Convert experiments into production-grade Python/FastAPI services; containerize with Docker and support Kubernetes/GPU deployment.
12
Optimize inference latency, GPU memory and throughput for enterprise NVIDIA infrastructure such as H100/A100 or equivalent.
Technology stack
Tools & Technologies.
The platforms and technologies you'll use to build modern, enterprise-grade solutions.
01
Python
Required
02
SQL
Required
03
OpenCV
Required
04
YOLO
Required
05
Vision Transformers
Required
06
PaddleOCR
Required
07
Docling
Required
08
PyTorch
Required
09
TensorFlow
Required
10
LLM
Required
11
VLM
Required
12
prompt engineering
Required
13
tool calling
Required
14
function calling
Required
15
JSON structured output
Required
16
RAG
Required
17
Graph-RAG
Required
18
embeddings
Required
19
vector search
Required
20
hybrid retrieval
Required
Requirements
Skills & Experience.
We value curiosity, collaboration and continuous learning. If you don't meet every requirement but believe you can make an impact, we'd still love to hear from you.
Essential
Required Qualifications
Python
SQL
Computer Vision (OpenCV)
Object detection (YOLO or equivalent)
Segmentation
Layout understanding / document layout analysis
OCR / Document AI (PaddleOCR, Docling or equivalent)
Deep learning with PyTorch and/or TensorFlow
Model evaluation (precision/recall/F1, IoU, OCR CER/WER)
LLMs/VLMs (multimodal), prompt engineering
Structured outputs (JSON) and guardrails
RAG fundamentals (embeddings, chunking, vector search, hybrid retrieval, reranking, query transformation)
Knowledge Graphs / graph databases (Neo4j or equivalent)
Graph retrieval / Graph-RAG concepts
LLM optimization (LoRA/QLoRA, PEFT, quantization, fine-tuning, context/inference optimization)
FastAPI/REST services
Docker
Kubernetes
Git
GPU deployment/optimization (latency, memory, throughput)
Preferred
Nice to Have
Experience with engineering drawing automation / technical document intelligence
Domain exposure: aerospace, mechanical/manufacturing, automotive, rail, CAD/PLM
GD&T knowledge
ASME Y14.x familiarity
ISO GPS familiarity
Experience operating in secure enterprise/on-prem environments with restricted internet access
Experience with NVIDIA enterprise GPUs (H100/A100 or equivalent)
What you'll gain
More Than Just A Job.
We're committed to helping every team member grow professionally, personally and technically while working on meaningful projects.
Global Exposure
Collaborate with international clients and multicultural teams on enterprise programmes.
Continuous Learning
Expand your expertise through mentoring, certifications and hands-on project experience.
Career Growth
Take ownership, develop leadership skills and grow your consulting career over time.
Flexible Working
Hybrid and remote collaboration designed around trust and delivering exceptional outcomes.
People First
Join a supportive culture where collaboration, respect and long-term relationships come first.
Enterprise Projects
Work on meaningful technology initiatives for leading organisations across industries.
Apply
Apply for Senior Data Scientist / AI Engineer (Computer Vision, OCR/Document AI, Multimodal LLM, RAG/Graph-RAG)
One page, about two minutes. We only ask for what we actually need to have a first conversation.