In a nutshell 

Picnic is building a new generation of intelligent shopping experiences and fulfillment automation with AI. In our app, we help customers search in their own words, plan meals around their household and shop effortlessly. In our supply chain, automation helps us deliver efficiently and sustainably. 

As an AI Platform Engineer, you’ll build the tools and infrastructure to take AI from experiment to dependable production system: model access, tracing, evaluation, fine-tuning and low-latency serving. You'll set the technical direction for how engineers build with AI, working closely with product teams across search, personalisation, and fulfillment automation.

Tricks of the trade

Provide an environment in which agents run at optimal cost: Evolve our AI Gateway with provider integrations, model routing, fallbacks, rate limits, privacy controls, and cost attribution.

Standardise LLM tracing, evals and benchmarking: Build end-to-end tracing, evaluation, and production monitoring across prompts, models, retrieval, tools, and agent state. Help teams turn human judgement and customer outcomes into repeatable evaluation sets.

Productionize open-weight models: Create the path to fine-tune, evaluate and serve open-weight models. Support real-time inference and high-throughput batch workloads, and make latency, performance and cost trade-offs visible.

Build a paved path for getting agents to run in production: Enable developers by providing clear APIs, SDKs, templates and local tooling for structured generation, tool use, model experiments and agent workflows. Make the supported path easier than assembling one-off infrastructure.

Our Tech: You’ll work with Python on a daily basis, using frameworks like FastAPI, LiteLLM, OpenAI, Gemini and more. Supporting this core stack are Snowflake, PostgreSQL, OpenSearch, RabbitMQ, Kafka, Git, Docker, Kubernetes, AWS, Helm, Terraform, Vault, and Datadog. Our open-weight training and serving stack is one of the choices you'll help us make.

We’re looking for

• Bachelor’s degree or higher in Computer Science, AI or a related field
• Strong backend engineering fundamentals, especially in Python
• Proven track record building self-service AI platforms and scaling their adoption
• Experience bringing language models or other ML/AI systems into production with meaningful reliability, latency, security, and cost requirements
• Familiarity with cloud infrastructure, containers, Kubernetes, CI/CD and observability
• Able to take end-to-end ownership, make pragmatic build-versus-buy decisions and work effectively across platform, ML/AI and product teams