document understanding, LLMs, multimodal AI, and agentic AI systems
.
You will lead the development of a
co-pilot-style solution
capable of ingesting, processing, and reasoning over large volumes of documents (PDFs, scanned images, engineering drawings, invoices, P&IDs, SOPs, etc.), and delivering accurate, context-aware answers to user queries.
This role requires deep expertise in both
cloud-native (Azure)
and
open-source AI ecosystems
, with the ability to design
autonomous, agent-driven workflows
that integrate multiple tools, APIs, and data sources.
(using LangChain Agents, Semantic Kernel, AutoGen, etc.) for multi-step reasoning and automation.
Integrate
OCR, computer vision, and NLP models
to process scanned PDFs, engineering drawings, invoices, and P&ID diagrams.
Fine-tune and optimize LLMs (OpenAI, Azure OpenAI, Hugging Face, LLaMA, etc.) for enterprise use.
(e.g., Pinecone, Weaviate, Azure AI Search, FAISS) for semantic and multimodal search.
Design and manage efficient
data pipelines
using Python, SQL, and distributed processing frameworks.
Full-Stack Collaboration
Work with front-end developers (React) to integrate AI services into user-friendly applications.
Ensure smooth API design and deployment of AI services.
Performance & Governance
Ensure system scalability, low latency, and cost efficiency.
Apply best practices for
responsible AI
(bias mitigation, data security, compliance).
Required Skills & Qualifications
-------------------------------------
Core AI/ML Expertise
Strong hands-on experience in
NLP, LLMs, RAG, and multimodal AI
.
Proven experience building
agentic AI workflows
(LangChain Agents, Microsoft Semantic Kernel, AutoGen, CrewAI, Haystack agents).
Proficiency with
document AI, OCR, computer vision
, and structured document parsing.
Expertise in
transformer-based models
(BERT, GPT, LLaMA, etc.), fine-tuning, and prompt engineering.
Knowledge of