As an AI/ML Engineer at KadamTech, you will build AI agents, RAG pipelines, and intelligent automation that solve concrete business problems rather than demos that never leave a notebook. You will work primarily in Python, integrating large language models, embedding stores, and vector databases into products that customer-support teams, analysts, and operations staff actually use every day.
The problems are diverse and practical. You might build a retrieval-augmented assistant over a client's knowledge base, design an automation that classifies and routes thousands of documents, or wire up an agentic workflow that takes actions across multiple tools and APIs. A big part of the job is rigorous evaluation: defining what "good" means for a given task, building test sets, measuring quality and cost, and tuning prompts, retrieval, and models until the system is genuinely reliable.
This role is exciting because you sit at the intersection of fast-moving research and real production engineering. You will own features from prototype to deployment, make pragmatic calls on build-versus-buy, and put guardrails around models so they behave safely and predictably. If you like shipping AI that earns its keep, you will have a lot of room to make an impact here.
What you'll do
- Build and ship LLM-powered features such as assistants, summarisers, and classifiers
- Design and implement RAG pipelines, including chunking, embeddings, and retrieval strategies
- Develop agentic and automation workflows that integrate multiple tools and APIs
- Define evaluation criteria and build test sets to measure model quality, latency, and cost
- Optimise prompts, retrieval, and model selection to improve accuracy and reduce spend
- Deploy models and pipelines to production with monitoring, logging, and safety guardrails
- Collaborate with engineering and clients to scope realistic, high-value AI use cases
What we're looking for
- 3+ years of software engineering experience, with strong production-grade Python skills
- Hands-on experience with LLM APIs (e.g. OpenAI, Anthropic) and prompt engineering
- Experience building RAG systems and working with vector databases (e.g. pgvector, Pinecone, Weaviate)
- Working understanding of MLOps basics: deployment, monitoring, and versioning
- A rigorous, evaluation-driven approach to measuring and improving model quality
- Pragmatic, outcome-focused mindset that prioritises shipping reliable solutions
- Nice to have: experience with fine-tuning, orchestration frameworks (LangChain, LlamaIndex), or cloud ML services