AI Engineer

Role Overview

We are seeking an AI Engineer specializing in Natural Language Processing (NLP), Large Language Models (LLMs), and document intelligence for our Tender Management Solution (TMS). This role involves transforming prototypes into reliable product features that help enterprise teams analyze complex tender documents, identify bid requirements, assess eligibility, and leverage company knowledge to prepare stronger bids. You will solve advanced information-retrieval problems rather than building generic chatbots.

Core Responsibilities

• NLP & LLM Focus:
• Build NLP/LLM pipelines for processing lengthy government and enterprise tender documents (RFPs, RFQs, financial clauses, etc.).
• Create information-extraction workflows with validation and traceability to source text.
• Design Retrieval-Augmented Generation (RAG) systems that allow users to query company knowledge with evidence-based answers.
• Develop robust document ingestion strategies for PDFs, scanned images, tables, and multi-column layouts using OCR and layout-aware extraction.
• Implement semantic search, embeddings, reranking, and hybrid retrieval for relevant clauses and past credentials.
• Build AI-assisted eligibility checking by comparing tender criteria against company profiles (financial thresholds, certifications, experience).
• Create the TMS Company Vault: convert unstructured data into a searchable knowledge base of projects, CVs, and bid history.
• Develop workflows to match personnel qualifications and project history to specific tender requirements.
• Generate structured outputs like compliance matrices and risk flags with clear evidence citations.
• Evaluate LLM output quality (accuracy, groundedness) using test datasets and regression checks to minimize hallucinations.
• Experiment with prompt engineering, model selection, fine-tuning, and hybrid approaches balancing cost, latency, and performance.
• Create safeguards for sensitive data, low-confidence outputs, and high-impact decision approvals.

• Computer Vision:
• Support infrastructure monitoring (solid waste management) using OpenCV and detection models like YOLO.
• Annotate data, perform augmentation, evaluate models, and deploy solutions for real-world camera feeds in production dashboards.

Requirements

• Strong Python skills with practical experience building AI/ML applications.
• Deep understanding of NLP fundamentals; hands-on proficiency with LLMs, prompt engineering, or RAG systems.
• Familiarity with major model ecosystems (OpenAI, Gemini, Anthropic, Hugging Face).
• Experience with vector databases and semantic search tools (FAISS, Qdrant, Pinecone, Milvus, pgvector).
• Solid grasp of document processing: OCR, text extraction, chunking, metadata management.
• Basic knowledge of ML fundamentals and computer vision concepts (OpenCV, YOLO).
• Comfortable with APIs, Git/GitHub, debugging, and collaborating on software integration.
• Preferred experience with LangChain, LlamaIndex, or Hugging Face Transformers.
• Hands-on work with scanned PDFs, OCR engines, table extraction, and multilingual documents.
• Experience in RAG evaluation, building test datasets, reranking, hybrid search, or observability tools.
• Exposure to PyTorch/TensorFlow, image augmentation, or real-time video analytics is a plus.
• Familiarity with deploying services on AWS, GCP, Azure, or Linux (Docker experience preferred).
• Interest in enterprise workflows, bid/tender processes, and infrastructure consulting.

Candidate Profile

We value builders who enjoy researching problems and delivering solutions. You should have a product-first mindset focused on accuracy and explainability, prioritizing quality to handle exceptions and signal uncertainty rather than just generating fluent text. Curiosity and ownership are essential—you must be able to document decisions and collaborate closely with our tech lead.

Location