01
Kittl Agentic AI
Current hands-on production experience: owned the Agentic AI project at Kittl, from architecture and technical direction through to production delivery.
AI ENGINEERING SYSTEMS
Make AI-assisted software development work as a system.
Coding agents can make implementation dramatically faster. The harder question is what happens around them: context, architecture, verification, review, feedback loops and human judgment.
LAUENSTEIN One designs and improves those engineering systems so teams can delegate more work to AI without simply moving the bottleneck somewhere else.
START WITH YOUR BOTTLENECK
Choose the situation closest to your team.
AI-assisted engineering creates different bottlenecks at different stages. Start with the problem you already feel — not with another tool.
01 · SMALL AI-NATIVE TEAM
Our coding agents are useful, but they still need too much senior guidance.
Context, corrections, checks and agent direction still live in the heads of founders or senior engineers. Turn one recurring workflow into a repeatable delivery system.
Build a repeatable agent workflow02 · GROWING ENGINEERING TEAMAI makes implementation faster, but the rest of our delivery workflow is not keeping up.
The saved time reappears in requirements, context, architecture iterations, rework, verification, review or coordination. Find and fix the next bottleneck.
Improve delivery workflow03 · MATURE AI-ASSISTED TEAMOur bottleneck is now verification, review readiness and senior attention.
Implementation is fast enough. The expensive part is becoming confident that an AI-assisted change is ready for human sign-off.
Reduce review costNot sure which layer is actually slowing you down? Describe the workflow and we will help identify the right starting point.
PRODUCTION AI ENGINEERING
Production AI engineering — not prompt demos.
02
AI-assisted technical sign-off
Designed an AI-assisted engineering sign-off workflow adopted across approximately five product teams and used on 23 projects.
03
AI Change Verification — open source
We built and open-sourced an evidence-accountable verification workflow for AI-assisted code changes before human review — with deterministic evidence, unresolved-risk handling and explicit human judgment.
AI Change Verification on GitHubWORKING DIRECTLY WITH ENGINEERS
The engineers doing the work
AI & Software Engineering Lead
Yaroslava Suspitsina
Senior software engineer focused on production AI systems, agentic architecture and engineering workflows. She currently works at Kittl, where she owned the Agentic AI project from architecture and technical direction through to production delivery. She has also designed AI-assisted review, testing and quality workflows used across multiple product teams.
- Current Kittl Agentic AI project ownership and production delivery.
- AI-assisted technical sign-off used across approximately five product teams and 23 projects.
- Agentic architecture, testing, CI quality automation, Playwright, debugging and engineering enablement.
AI/ML Engineer
Valerii Safonov
AI/ML engineer with experience across machine learning, quantitative modelling, NLP/RAG, computer vision and production-oriented AI systems. MSc in Computer Science with a machine-learning focus, with applied AI engineering experience dating back to at least 2020.
- MSc in Computer Science with a machine-learning focus.
- Applied AI/ML work across quantitative modelling, NLP/RAG, computer vision and production engineering.
- Earlier applied technology-trend detection and evaluation work spanning data, architecture and engineering methods.
No handoff from sales to a junior delivery team. The people shown here work on the engagement.
THE SYSTEM AROUND THE MODEL
The model is only one part of the engineering system.
Better coding agents do not automatically create better delivery. As implementation gets faster, the surrounding engineering system matters more: the context agents can inspect, the constraints they can follow, the checks they can run, the feedback they can consume and the decisions that still belong to people.
We work on that layer — from agent-ready repository context and repeatable workflows to deterministic verification, review readiness and clear human decision boundaries.
We are tool-neutral. We do not sell generic AI training, broad “AI transformation” programmes or unnecessary platform migrations.
01
Context that agents can actually use
Repository knowledge, task intent, architecture and constraints should be inspectable rather than repeatedly explained from memory.
02
Repeatable agent workflows
Useful individual practices should become maintainable engineering workflows where repetition creates leverage.
03
Evidence before expensive human attention
Cheap deterministic checks and relevant evidence should happen before senior judgment wherever possible.
04
Humans own consequential decisions
Architecture, product intent, exceptions and genuinely ambiguous trade-offs remain explicit human responsibilities.
HOW WE WORK
One bounded engineering problem at a time.
No ongoing retainer is required. Further work only happens if there is a separate reason to do it.
01
Start with the real workflow
We scope one production workflow where human time is still being spent repeatedly around AI-assisted development.
02
Find the actual constraint
We baseline the current workflow and separate the real bottleneck from assumptions about what AI should improve.
03
Change the system, not the slide deck
We implement the smallest useful intervention in the real engineering environment and test it on live work.
04
Leave evidence and ownership behind
The engagement ends with a working change, before/after evidence, unresolved items and a handoff your team can maintain.
CONTACT
Show us where AI-assisted delivery still costs human attention.
Briefly describe your team, the AI or coding tools you already use, and the part of the workflow that still feels expensive, repetitive or uncertain.
We will tell you which Sprint appears closest to the problem — or if we do not think one is a good fit.