What You’ll Be Doing Design and prototype agent
workflows using LangChain and/or similar frameworks. Develop task
automation agents focused on marketing workflows (e.g., campaign
execution, reporting, content validation). Build basic RAG
pipelines and ReAct agents to improve agent response relevance and
accuracy. Benchmark and test LLMs for speed, cost, and fitness to
the use case. Work closely with the Hiring Manager To iterate on
use cases and validate the POC. Possible Extensions (Post-POC /
Platform Build) Scale agent orchestration using LangGraph, AutoGen,
or CrewAI. Containerize workflows and prepare deployment pipelines.
Integrate monitoring and observability layers for agent evaluation.
What You Need to Succeed Must-haves 3+ years of experience in
software engineering or machine learning, with at least 1 year of
hands-on experience building LLM-based workflows or autonomous
agents. Proven ability to design, prototype, and iterate on
solutions using LangChain or similar frameworks. Strong
understanding of prompt engineering, tool-use chaining, and agent




orchestration patterns (e.g., ReAct, Tool-Calling). Practical
experience working with vector databases such as Pinecone, FAISS,
or Weaviate, and building Retrieval-Augmented Generation (RAG)
pipelines. Proficiency in Python and relevant ML libraries (e.g.,
Transformers, PyTorch, LangChain). Familiarity with Docker and
experience deploying ML workloads in cloud environments (AWS, GCP,
or Azure). Ability to work independently, communicate effectively
with non-technical stakeholders, and iterate rapidly in a
PoC-driven environment. Nice-to-haves Experience with multi-agent
frameworks like LangGraph, AutoGen, or CrewAI. Previous work on
agentic systems applied to marketing automation or related use
cases. Knowledge of CI/CD practices, container orchestration, and
performance monitoring for ML pipelines. Awareness of AI ethics,
security, and compliance in production-grade applications.
Demonstrated contributions to

📌 AI Engineer (Cundinamarca)
🏢 Pearster
📍 Cundinamarca