Associate GenAI Engineer

Introduction :

Reconect.ai is building autonomous AI Workforce for Debt collection. Our platform empowers AI agents that don’t just talk — they think, plan, and act. These agents understand data, predict intent, choose the right strategies, and execute personalized communication flows across channels — all while staying compliant.

We integrate LLMs, event-driven orchestration, multilingual response handling, and contextual decision-making to transform how collections are planned and executed. As a GenAI / ML Engineer, you’ll design how our agents reason, respond, and adapt — by developing prompts, finetuning models, and optimizing behavior through routing and memory — making them more reliable, responsive, and effective with each interaction.

Responsibilities :

• LLM Prompt Engineering & Optimization : Design and refine system prompts, few-shot examples, and dynamic flows tailored for multilingual debt collection scenarios.
• Open-Source Model Finetuning : Adapt LLMs (e.g., LLaMA, Mistral) using LoRA, DPO, or SFT techniques on domain-specific data like chat transcripts and disposition labels.
• Model Evaluation & Behavior Control : Define and run evaluations across metrics like accuracy, tone, latency, and compliance. Tune models to behave predictably in production.
• RAG & Embedding Systems : Build RAG pipelines and maintain vector DBs to give agents contextual awareness, memory, and grounding in past customer history.
• Multilingual & Language Handling : Build language-aware agents capable of seamless code-switching and context-appropriate responses across Hindi, English, and other Indian languages.

Qualifications :

Why Join Us?