Opportunity at
FoodicsAI Engineer
Foodics is hiring an AI Engineer to build backend foundations for agentic AI, in which language models plan, act and invoke tools with minimal human steering.…
Career Tools
About This Role
Foodics is hiring an AI Engineer to build backend foundations for agentic AI, in which language models plan, act and invoke tools with minimal human steering. The emphasis is on infrastructure: orchestration, service interfaces, state handling, tracing and evaluation harnesses, delivered as dependable and scalable production systems. The job is practical backend engineering that centers on writing, testing and diagnosing code. Prompt design is included, but the heavier architecture work…
Job Description
Foodics is hiring an AI Engineer to build backend foundations for agentic AI, in which language models plan, act and invoke tools with minimal human steering. The emphasis is on infrastructure: orchestration, service interfaces, state handling, tracing and evaluation harnesses, delivered as dependable and scalable production systems. The job is practical backend engineering that centers on writing, testing and diagnosing code. Prompt design is included, but the heavier architecture work sits in the infrastructure layer.
Candidates at all levels are considered, from early-career developers to veteran architects. Newer engineers are mentored by seniors and are expected to take ownership of production components inside three months, while senior engineers set technical standards and direction. Where someone lands depends on demonstrated capability.
Typical work involves tool integrations and Model Context Protocol servers with robust error recovery, end-to-end tracing of model calls, retrieval and tool runs, evaluation suites and regression tests, and root-cause analysis of loops, context decay and invented outputs. The team also enforces guardrails, spending limits, role-based access, output validation and human sign-off triggers. Work follows OpenTelemetry GenAI conventions and OWASP guidance for agentic systems. The posting says prior work with specific agent frameworks, academic AI research and base-model training are not necessary. The role is listed on-site and mentions hybrid arrangements. It is full-time, and no salary is given.
Responsibilities
- Construct agent infrastructure covering orchestration, sessions, state and backend service interfaces
- Design tool connectors and MCP servers with clearly bounded responsibilities and error recovery
- Add tracing and observability to model inference, document lookups and tool runs
- Build test harnesses, scoring sets and regression checks that cover real production edge cases
- Diagnose runtime failures, loops, context degradation and hallucinations to harden the system
- Implement guardrails, cost ceilings, permission controls, output schemas and human review triggers
- Track research and architecture developments and share them with engineers
Requirements
- Strong software engineering in a modern language, with algorithmic depth
- Grounding in data structures, algorithms, memory management and concurrency
- Backend skills with REST or gRPC APIs, databases, caches, queues and resilient distributed design
- Competence with Git, automated tests, CI/CD and maintainable code
- Working model of LLM internals: context limits, tokenization, function calling, non-determinism and hallucination control
- Completed full-lifecycle projects, whether in production, coursework, open source or personal builds
- Methodical debugging and clear technical writing
Qualifications
- Built prototypes on LLM APIs
- Experience with MCP or bespoke tool connectors
- Knowledge of vector indexes, embeddings and retrieval-augmented generation
- Cloud (AWS preferred), containers (Docker, Kubernetes), telemetry or infrastructure-as-code
- Grasp of GenAI security, prompt injection defense and threat modeling
- Open-source contributions or community involvement
Core Skills
Benefits
- Pay with bonuses and possible share participation
- Ongoing training and a yearly learning allowance
- Independence, mentoring and ambitious goals
- Flexible in-office and hybrid arrangements
Frequently Asked Questions
Answers are based only on the employer’s listing; where it doesn’t say, neither do we.
Is prior agent-framework experience required?
No. The posting says extensive experience with particular agent frameworks is not required, and early-career candidates are welcome.
What is the work arrangement?
The listing is on-site, and the posting mentions hybrid setups. No city is named.
What do junior engineers do early on?
They pair with senior mentors and are expected to take ownership of production components inside three months.
What is the salary?
No salary is given.
Sample Interview Questions
AI-generated questions tailored to this specific role — a preview of the full practice set.
