AI Full-Stack Developer, New Grad

About Concurrences

Mission: "Connecting Antitrust People." The global platform for competition law — 20+ years, 150+ countries, ~50,000 professionals. Lawyers, economists, judges, regulators, academics: all shaping competition policy. We started as a challenger taking on established players. That DNA hasn't changed — now we're building with generative AI.

What drives us: Excellence — scientific rigor and editorial quality in everything we publish. Innovation as a driving force — pioneering new formats, anticipating industry trends, now building with generative AI. And a global community connected through high-level dialogue and best practices shared across borders.

The Role

You own the RAG and inference infrastructure for the world's first production AI built for antitrust law. October 2026 launch, 50+ competition authorities worldwide. You're not junior support — you own systems end-to-end: retrieval pipelines, vector databases, inference orchestration, API design. You make architecture decisions. You ship every week.

Why this matters: Concurrences.com is the definitive antitrust platform (20 years, 100K+ articles). Now we're turning 330–430K legal documents into a production-grade agentic research engine.

What You'll Build

• RAG pipelines: chunking, embedding strategies, retrieval optimization (BM25 + semantic + reranking), latency tuning

• Python + FastAPI APIs: exposing search, retrieval, agentic workflows to Claude, ChatGPT, enterprise copilots

• Weaviate: schema design, indexing strategy, query optimization, cost-per-query tuning

• LLM integration: managing context, prompt templates, cost/latency trade-offs

• Agentic systems: CrewAI, LangGraph, or LlamaIndex for multi-step research flows

• Infrastructure: CI/CD with Pulumi/Terraform, monitoring, observability

A Typical Week

• Monday: Bug in chunking pipeline. Root cause, fix, PR, test. Partner with LLM engineer on retrieval validation.

• Tuesday: Customer request: search across EU + US jurisdictions. Redesign query routing.

• Wednesday: Code review: intern's reranking module is slow. Pair on profiling, async batching.

• Thursday: Monitoring alert: 2-second tail latency. Design two-tier indexing. Draft design doc.

• Friday: Deploy indexing strategy, monitor post-launch. Async update to Anton.

Who You'll Work With

Anton Bossenbroek (CTO): 20+ years at BCG, Bain, EY, Deloitte, and C3.ai. Your primary technical collaborator.

Nicolas Charbit (founder): 20 years building Concurrences into the definitive antitrust platform.

Small team. Everyone ships. High signal-to-noise ratio.

Must-Have Qualifications

• Strong Python — ship production code daily. FastAPI is a big plus.

• Built APIs or backend systems — something other code calls.

• Hands-on LLM APIs — you've integrated Claude, ChatGPT, or another LLM into a system.

• Git and DevOps thinking — version control, CI/CD concepts, deploying code.

• Production mindset — you think latency, cost, monitoring, failure modes.

Big Pluses

• Vector database hands-on: Weaviate, Pinecone, pgvector, FAISS, or Milvus

• Agentic frameworks: CrewAI, LangGraph, LlamaIndex, or LangChain

• Infrastructure as Code: Terraform, Pulumi, or CloudFormation

• GCP experience: Cloud Run, BigQuery, Pub/Sub, Cloud Storage

• FastAPI specifically: production FastAPI systems

• Law background: law degree, coursework, antitrust thesis

Compensation & Logistics

R$6,600–R$11,000/month (~€1,200–€2,000), paid in EUR via Wise. Independent contractor, monthly invoicing.

Location: Remote-first, async work culture. The CTO is based in Vancouver, the company in Paris. English required, French a plus. Start: ASAP, within 4 weeks.

Interview Process

1. Application — CV/LinkedIn, GitHub link, 4 qualifying questions (submitted via application form below).

2. 15-min intro call with Anton (CTO) — Why this role? Culture fit and motivation.

3. Two 30-Minute Deep Dives — System design and real problem solving.

4. 20-min conversation with Nicolas (founder) — Domain understanding, alignment.

Growth Path

Phase 1: Component Builder → Phase 2: Pipeline Owner → Phase 3: System Architect → Phase 4: Technical Lead or Product Engineer.

Originally posted on Himalayas