Objective
- Design and prototype an autonomous agent that performs continuous commercial watch and market intelligence.
- Detect relevant market signals and generate potential leads or alerts tailored for the sales team (prioritization by relevance, urgency, strategic fit).
Key features / Responsibilities
- Daily scanning of open web sources: news, tenders, social media, competitor sites, RSS feeds, APIs, and LinkedIn/Twitter if accessible.
- Extraction of relevant information: keywords, named entities, opportunities, trends, and events useful for lead detection.
- Scoring and prioritization of insights: implement relevance, urgency and strategic-fit metrics to surface high-value leads.
- Report generation and UI: produce periodic reports and/or live dashboard updates; optional chatbot/assistant interface to interact with the agent.
Technological environment & architecture
- Languages & frameworks: Python with Scrapy and Asyncio; use LangChain or CrewAI for agent orchestration.
- NLP & models: integration with OpenAI / HuggingFace models and spaCy for NER and text processing.
- Data handling & tools: web scraping, RSS/APIs ingestion, ElasticSearch for indexing, Streamlit or Dash for front-end dashboards; containerized with Docker.
- Architecture: event-driven micro-agent (async) architecture to enable continuous, scalable monitoring and modular agent behaviors.
Expected deliverables & evaluation
- A working prototype of the autonomous agent performing continuous monitoring and producing prioritized signals/leads.
- Source code, deployment artifacts (Docker), and a demonstration dashboard or sample periodic report.
- Documentation covering data sources, scoring/prioritization logic, model choices and setup instructions; evaluation based on detection accuracy, prioritization quality and system robustness.
How to apply
- Apply via the company website: https://www.hydatis.com
- Or send your application by email to stages@hydatis.fr with the subject: "AI Agents-1 - AI Agent for Autonomous Market Intelligence & Lead Detection".