About the role
You will own the AI/ML core for an AI-powered sales intelligence platform that helps enterprise sales teams close more deals. You will work on building the AI layer that gives sales representatives the right sales-action signals at the right moment.
You will be responsible for the RAG systems that power briefings, data pipelines that process real-world sales data, and the scoring models that assess deal health. You will apply your expertise to understand RAG at a fundamental level, not just how to call an API, but how these systems actually work and how to fix them when they break.
This position is in a pre-seed, fast-moving startup environment. You should be ready for ambiguity, rapid iteration, and assuming wider ownership, as you will be one of our first engineering hires building something reps say they’d pay for tomorrow.
Responsibilities
Build production RAG systems that ingest, chunk, embed, retrieve, and synthesize across heterogeneous data sources (CRM records, call transcripts, emails, docs).
Develop a deal health scoring model grounded in real historical data.
Design prompt engineering and LLM orchestration pipelines that are reliable, cost-efficient, and auditable.
Create data cleaning and normalization logic for dirty, inconsistent enterprise data.
Build evaluation frameworks to catch score/reality drift before it erodes customer trust.
Develop the five-pillar competitive intelligence algorithm: buyer priorities, competitor vulnerabilities, company strengths, proof points, and deal-momentum signals.
Collaborate directly with product-obsessed founders to shape the AI architecture from the ground up.
Main Requirements
Proven expertise in ML/NLP/information retrieval.
Strong ML & AI fundamentals: you understand how RAG systems are built at a low level, not just how to use LangChain.
Experience building and debugging production RAG or LLM pipelines (not just prototypes).
Experienced with vector databases, embedding models, retrieval strategies, and reranking.
Python fluency
Experience with LLM APIs (OpenAI, Anthropic, etc.).
Skilled in working with messy real-world data.
Skilled in diagnosing data retrieval failures and fixing them.
Self-learner, comfortable with AI tools and attention to detail.
Other Requirements
Advanced English.
Nice to Have, but Not Mandatory
Background in sales tech, CRM data, or enterprise SaaS data models.
Experience with multi-agent systems or LLM orchestration frameworks.
Experience building scoring or ranking systems.
