Location not specified
Published on: November 13, 2026
About Our Client
Our client is an AI-powered enterprise platform that automates end-to-end recruitment and workflow orchestration. The platform runs at production scale: it coordinates complex third-party integrations, processes large volumes of unstructured data, and runs event-driven pipelines across multiple systems in real time.
We're looking for a Senior Back-End Engineer to own the platform's Python microservices, data pipelines, and agents stack. You'll turn raw, messy data from many sources into clean, structured records and search indices that power AI agents. You'll join a small, high-ownership team where engineers drive decisions from pipeline architecture to cloud infrastructure, and ship independently without heavy process. The team works 100% in English.
Design and operate high-throughput data pipelines in Python and FastAPI
Build extraction, parsing, and enrichment workflows that turn raw files and feeds into structured, searchable records
Design and maintain Elasticsearch/OpenSearch indices, from mappings to bulk ingestion and query performance
Develop the runtimes AI agents run on: tool calling, execution state, context limits, and streaming
Integrate GraphQL, internal, and external vendor APIs with solid error handling and rate limiting
Build resilient event-driven systems on AWS (SQS, ECS, S3, RDS) with idempotency, retries, and dead-letter handling
Fluent English (C1 or higher), spoken and written
Deep, production-grade experience with Python and FastAPI, including async services and clean API contracts
Hands-on experience building pipelines that ingest and transform large volumes of messy or semi-structured data
Practical Elasticsearch or OpenSearch experience: index design, mappings, bulk indexing, and query optimisation
Working experience with GraphQL APIs and cross-service API contracts
Strong AWS knowledge (ECS, SQS, S3, RDS) and PostgreSQL skills, including async ORMs like SQLAlchemy
Experience with message queues, retries with back-off, idempotency, and dead-letter queues
The ability to own system design end to end, and to use logs, traces, and metrics to diagnose live issues
Clear written communication, a considered point of view in design discussions, and a bias toward simplicity
LLM integrations and agent workflows (LangGraph, LangChain, LiteLLM, tool calling)
TypeScript / Node.js (NestJS)
Parsing complex documents (PDF, DOCX, large spreadsheets)
Infrastructure as code (AWS CDK, Terraform)
Workflow orchestration tools such as Temporal or Inngest
This posting may involve the use of AI tools to assist in reviewing applications; final hiring decisions are made by humans.
As part of our hiring process, this role may use Keystone AI’s artificial intelligence or automated tools to assist with reviewing and screening applications. These tools support, but do not replace, human judgment in making hiring decisions.
Your application will be considered complete only after you have completed the full registration process on the Keystone AI platform, including creating your profile, uploading your CV, and submitting your application. The AI interview is optional but encouraged and is not required for your application to be considered.
By submitting your personal data and/or resume, you give consent to collection, use and disclosure of your personal data and/or resume by the company (or its agent) for the purpose of the processing and administration by the company relating to this job application. Read our Privacy Policy.
CGP Group is an affirmative equal-opportunity employer and recruitment firm. We evaluate qualified applicants without regard to race, color, religion, creed, gender, sexual orientation, gender identity, marital status, national origin, age, veteran status, disability, or any other protected class. Information provided in this advertisement is for recruitment purposes only.