ML / LLM Engineer – AI Agents

Bitmanager.ai

Bitmanager.ai is an AI startup in one of the fastest-growing categories of conversational AI agents.

We are currently strengthening our Agents stream and are looking for an experienced ML / LLM Engineer who can and loves to quickly bring AI agents from prototype to production.

What we are building

We are creating a platform for launching voice and text AI agents in customer communications: sales, support, consultations, and debt collection.

The agent interacts with the client, works with data, invokes tools, accesses internal company systems, and acts within a defined business scenario.

The platform allows for rapid configuration of agents for different processes and their deployment, taking into account the client's security, load, and infrastructure requirements.

Why it's interesting

An AI agent for a demo can be assembled quickly. Bringing it to enterprise-level production is much more challenging.

We are building a system that:

• stably conducts dialogue in real business scenarios;
• adheres to defined constraints and guardrails;
• works with tools and the client's internal systems;
• is verified through an evaluation infrastructure;
• is optimized for latency, cost, and stability;
• withstands real load;
• can be deployed on-premise;
• delivers high quality with small models.

This is an engineering challenge for someone who is interested in building complex AI agents that easily withstand product constraints.

What you will be doing

• Developing LLM and NLP components for production systems.
• Designing the architecture of AI agents: tool use, orchestration, multi-agent workflows, memory/context, evaluation, guardrails.
• Working on real-time voice conversational scenarios and optimizing pipeline latency, throughput, cost, and stability.
• Integrating AI agents into the product platform, client business systems, and enterprise environments.
• Developing the evaluation infrastructure: datasets, experiments, quality metrics, and agent behavior analysis.
• Participating in on-premise deployment of ML / LLM services.
• Quickly verifying technical and product hypotheses with the team.

What's important to us

We need an engineer who can solve complex technical problems and influence product development while adhering to business constraints, stability, performance, and cost requirements.

It's important to be able to quickly learn new things, work hands-on, take responsibility, and be comfortable working at the pace of an early-stage startup.

We expect you to have:

• experience in developing production ML / LLM systems under real load;
• practical experience with AI agents / LLM agents / agent-based systems;
• understanding of tool use, function calling, orchestration, and agent workflows;
• understanding of LLMs: transformers, tokenization, fine-tuning, evaluation, inference;
• proficient Python and practical experience with PyTorch;
• experience with production serving for ML / LLM: APIs, queues, scaling, fault tolerance;
• engineering mindset and the ability to see the system as a whole;
• readiness to actively use AI tools in your work.

If you enjoy speed, short cycles, and the opportunity to directly influence the product, you will be comfortable with us.