Techery

Senior / Staff Applied AI Engineer

Techery  •  Kyiv, UA / Dnipro, UA (Onsite)  •  1 month ago
Apply
AI can make mistakes so check important info. Chat history is never stored.

Job Description

Senior / Staff Applied AI Engineer

Application Deadline: 31 August 2026

Department: Engineering

Employment Type: Full Time

Location: Kyiv, Dnipro

Since 2013, Techery has been bridging the gap between business vision and technology. As a product and consulting company, we focus on building our own products that are both technologically advanced and tailored to meet real-world needs. Innovation is at our core, and we are increasingly leveraging the power of AI to develop cutting-edge features and explore new possibilities within our product portfolio.
Beyond our product development, we act as a consulting software company for non-tech clients and are in full charge of solving their challenges. Whether it's addressing complex issues or turning innovation into reality, our commitment ensures that the best visions are fully realized, providing clients with comprehensive solutions that drive success.
Beacon is a core product within Treel OS — an AI-powered data analyst and sales-strategy assistant that connects to enterprise data, services, and workflows. Beacon goes beyond chat: it generates persistent dashboards, analytical views, reports, workflows, and on-demand interfaces tailored to a specific company, role, user, and objective. We're looking for a Senior or Staff-level engineer who can turn ambiguous AI product problems into production systems, validate them with evidence, and improve quality, latency, and cost through hands-on engineering.

Key Responsibilities

  • Design and implement agent execution workflows with structured, validated model outputs
  • Build semantic routing and complexity-classification systems across skills, data sources, and models
  • Develop classifiers using BERT-style models, embeddings, rules, and ensembles
  • Implement multi-turn context handling and state/context-retrieval mechanisms
  • Build model-independent interfaces across OpenAI, Google, and open-weight models
  • Create synthetic-data generation and validation pipelines
  • Fine-tune and evaluate small models and classifiers
  • Build grounding, provenance, and factuality/claim-validation mechanisms
  • Create automated offline and online evaluation systems
  • Analyze production failures and convert them into product improvements
  • Optimize quality, latency, and inference cost
  • Write production backend code and tests, contributing to architectural decisions through implementation

Skills, Knowledge and Expertise

  • Substantial experience building production backend systems under ambiguous product requirements
  • Hands-on experience implementing LLM-powered applications beyond prompt prototypes (structured outputs, tool execution, agent orchestration)
  • Experience with retrieval, classification, ranking, or semantic routing
  • Background in classical machine learning and modern language models
  • Experience with embeddings and vector-based systems
  • Familiarity with supervised fine-tuning, LoRA, adapters, or distillation
  • Experience with evaluation design and model error analysis
  • Experience with synthetic-data generation
  • Strong backend engineering skills in TypeScript, Rust, or another strongly engineered ecosystem
  • Practical, disciplined experience working with coding agents (codebase exploration, task decomposition, test generation, large refactors, failure investigation)
Techery

About Techery

Techery is a top software consultancy helping businesses achieve real technological modernization.

Our customers realize that the map is being changed and they need more advanced tools to execute their strategies.

Unlike many other digital companies Techery not only knows HOW to write code, but also knows WHAT to write and WHEN to write it. We are one of the few technology companies that won't ask questions like:

- "Now please tell us about your business"​

- "So what do you want to build?"​

- "OK, do you have requirements for the software you have in mind?"​

Instead we will ask questions like:

- "So what's on your mind? You go first and then we'll share what we were thinking as well"​

- "Is that true that your biggest challenge tomorrow is gonna be XXXXX?"​

- "What are ready to say NO to today in order to get what you want tomorrow?"​

Of course, like everyone else we have some big names up our sleeve, but there's no point in trying to impress anyone with them. However, our concept of working with large companies is different. We only work with them to deliver something truly groundbreaking that will eventually impact their entire business, like it was with Xerox and the CloudPrinting platform we had built.

We do love smaller companies more though. Better communication, less BS, straight goals, clear metrics. And that's understandable. Every big company is trying to make it look like it's small and to operate like it's small. We can help with that too.

We prefer personal meetings to endless chains of emails.

Added value to process following.

Innovation to Stagnation and Status Quo.

Reach out today and we promise we're gonna have a great time.

(and if our first meeting doesn't work out – at least we'll both enjoy our delicious dinner and interesting conversation! It's on us.)

Alex Pshenianykov, President

alex.p@techery.io

Industry
IT & Software
Company Size
51-200 employees
Headquarters
Plano, Texas
Year Founded
2007
Social Media