Tech Lead LLM Python APIs RAG Web - pré embauche client final possible - Freelance (H/F)

France | Sept. 26, 2026

Report as Closed

Company: Collective.work

Country: France

Type: Onsite

Employment: Full-time

Description: CONTEXT

PRE HIRE - 6 months maximum mission
You are joining the AI ​​Service team responsible for a strategic technological building block of the group's platform. International mission to provide AI capabilities to products used by legal professionals. As Techlead, you design robust APIs orchestrating LLM calls, RAG pipelines and massive processing, with streaming constraints, low latency, concurrency, parallelism, scalability, reliability and multi-tenant architecture (multiple countries).

MISSIONS

*

   Design and development of editorial data indexing and restitution consumption services:

   *

   Architecture APIs and workers, cache strategy, resilience mechanisms (retries, circuit breakers, fallback models).

   *

     LLM integration: inference chains (prompting, function calling, RAG, embeddings, agents), industrialization via POCs and benchmarks.

   *

     Performance and scalability: trade-off throughput, latency and cost (synchronous/asynchronous).

*

   Web APIs and performance (synchronous / asynchronous):

   *

     Definition of robust and documented APIs (FastAPI): contracts, validation (Pydantic), unified authentication and propagation of the country context.

   *

     Mastery of asyncio, concurrency, parallelism; implementation of LLM streaming, timeout management, backpressure and cancellation.

   *

     Observability and security: telemetry (logs, metrics, traces), secrets management, Zero Trust, token isolation.

*

   Technical leadership and governance:

   *

     Inner Source governance, squad support, promotion of Clean Code practices, SOLID, tests automated, monitoring and technical outreach.

TECHNICAL STACK

*

   Languages: Python 3.11+ (proficiency required), Go, Java / Spring Boot

*

   Web / APIs: FastAPI, Starlette, Pydantic, Uvicorn (ASGI)

*

   Asynchronous: asyncio, httpx, concurrency and parallelism (threads, processes, workers)

*

   LLM: OpenAI / Azure OpenAI, LangChain, Mistral, Bedrock, etc.

*

   Cloud: AWS, Azure

*

   Data & infra: OpenSearch, Weaviate, PostgreSQL, Redis, Celery, Kafka, Docker, CI/CD

*

  Tests: Pytest, pytest-asyncio

WORKING CONDITIONS

*

   Contract: pre-hiring, mission of 6 months maximum

*

   International environment, daily European context

*

   Remuneration: range to be defined

ADDITIONAL INFORMATION

Subscribe to the page NaxoTech on LinkedIn to discover our regular offers:
https://www.linkedin.com/company/naxotech (https://www.linkedin.com/company/naxotech)

1.  Diploma: Master/Engineer/Ph.D
2.  Proficiency required in Python 3.11+
3.  Knowledge of languages: Go, Java / Spring Boot
4.  Experience with Web frameworks / APIs: FastAPI, Starlette, Pydantic, Uvicorn (ASGI)
5.  Proficiency in asynchronous programming: asyncio, httpx, concurrency and parallelism management (threads, processes, workers)
6.  Experience in integrating language models (LLM) in production: OpenAI / Azure OpenAI, LangChain, Mistral, Bedrock, etc.
7.  Knowledge of Cloud environments: AWS, Azure
8.  Data and infrastructure skills: OpenSearch, Weaviate, PostgreSQL, Redis, Celery, Kafka, Docker, CI/CD
9.  Proficiency in testing: Pytest, pytest-asyncio
10. Experience in large-scale Python backend development
11. Proficiency in API contracts, versioning, dependency management and tokens
12. Ability to arbitrate between synchronous and asynchronous while guaranteeing performance and security
13. Fluent English (daily European context)

Apply here:

Web: Apply here

Emails: