AI / ML Engineer

Problem Identification and Solution Design:

• Understand business problems and design AI-driven automation solutions.
• Architect scalable systems combining ML models, LLMs, and rule-based logic.

Data Collection and Preprocessing:

• Collect, clean, and preprocess structured and unstructured data.
• Build pipelines for document ingestion, embeddings, and retrieval systems.

Model Development and Training:

• Develop and fine-tune ML, NLP, and Generative AI models.
• Work LLMs and SLMs (Small Language Models) for optimised use cases.
• Apply fine-tuning techniques (LoRA, PEFT) for efficient model adaptation.
• Implement embedding models, semantic search, and ranking systems.

RAG and Knowledge Systems:

• Design and implement RAG (Retrieval-Augmented Generation) pipelines.
• Work on vector databases and hybrid retrieval strategies.
• Build or knowledge graphs for enhanced reasoning.

Agentic AI and Orchestration:

• Build agent-based systems using LangChain, LangGraph, or similar frameworks.
• Design multi-agent workflows, tool usage, and orchestration pipelines.
• Implement agent capabilities, memory, planning, and reasoning loops.

Model Evaluation and Validation:

• Evaluate models' precision, recall, F1-score, and LLM-specific eval methods.
• Reduce hallucinations and improve response quality using prompt and system design.

Deployment and Integration:

• Build and deploy APIs with Flask / FastAPI.
• Integrate PostgreSQL and vector databases (FAISS, Pinecone, Chroma, etc. )
• Deploy cloud platforms (AWS/GCP/Azure) or on-prem/local environments.

Monitoring and Optimisation:

• Monitor performance (accuracy, latency, cost) and continuously improve systems.
• Optimise pipelines, prompts, and models for production readiness.

Ethical AI and Compliance:

• Ensure fairness, bias mitigation, and safe AI practices.
• Implement guardrails and compliance-aware AI systems.

Requirements:

• Strong proficiency in Python.
• Hands-on experience with ML frameworks (PyTorch / TensorFlow).
• Experience LLMs, SLMs, embeddings, and RAG pipelines.
• Strong understanding of fine-tuning techniques (LoRA, PEFT).
• Experience LangChain, LangGraph, or agent orchestration frameworks.
• Hands-on experience with Flask / FastAPI APIs.
• Strong knowledge of PostgreSQL and vector databases.
• Experience automation systems/decision engines / rule-based systems.

Good to Have:

• Experience MLOps practices and tools (CI/CD for ML, model versioning, monitoring).
• Familiarity with knowledge graphs (Neo4j, etc. )
• Experience local/on-prem LLM deployment and optimisation.
• Exposure to real-time/event-driven architectures.
  • Background in fintech/compliance/transaction monitoring systems.