2 months ago
Gurgaon, IndiaSenior
Responsibilities
- Build scalable AI services for prompt orchestration, model routing, embeddings, structured generation, tool calling, and multi-provider runtimes.
- Design multi-agent and autonomous systems with reasoning, planning, memory, tool execution, graph-based workflows, and human-in-the-loop checkpoints using MCP and A2A.
- Develop enterprise RAG pipelines covering ingestion, chunking, embeddings, hybrid search, reranking, citations, and retrieval-quality evaluation.
- Train, fine-tune, evaluate, and productionize ML and deep-learning models using techniques including LoRA, QLoRA, and PEFT.
- Deploy and operate agents and models with experiment tracking, model registries, versioning, rollback, drift monitoring, cost monitoring, and canary or blue-green deployments.
- Implement guardrails for hallucinations, prompt injection, and PII, along with responsible-AI evaluation and compliance practices.
- Build cloud-native AI services on GCP and Azure and own them through production support.
- Develop React interfaces for chat, copilots, prompt playgrounds, and agent or evaluation dashboards.
- Participate in architecture and code reviews, write tested code, and mentor engineers.
Requirements
- Bachelor’s or master’s degree in Computer Science, AI, Engineering, or a related field.
- At least 8 years of software engineering experience, including at least 3 years building production AI/ML applications.
- Strong grounding in distributed systems, cloud-native architecture, and microservices.
- Demonstrated experience delivering enterprise-grade AI solutions from proof of concept through production.
- Expert Python skills and experience with TypeScript or JavaScript and SQL.
- Experience with model training and fine-tuning, LLMs, RAG, agentic AI, hyperparameter optimization, GPU optimization, and quantization.
- Experience with AI frameworks, cloud AI platforms, ML/MLOps tooling, agent observability, vector databases, cloud infrastructure, and workflow orchestration.
Tech Stack
AzureDockerGitHub ActionsGoogle Cloud PlatformGrafanagRPCJavaScriptKubernetesMLflowPythonPyTorchReactSQLTensorFlowTerraformTypeScript
Categories
About dunnhumby
Dunnhumby builds customer data science and retail media software and services for retailers and consumer brands, supporting loyalty, pricing, and category decisions. It sells SaaS analytics platforms and consulting, and operates retail media solutions with partners. Founded in 1989 and headquartered in London, it is owned by Tesco and works with companies such as Tesco, Coca‑Cola, and Procter & Gamble.