The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Back to feedPost
I stopped hiring people to fix operational problems.
I took part in Anthropic's AI Interviewer study. One AI asked the questions; another helped me answer in English. The conversation became a summary of what I've been doing in my own business: not adding AI to the company, but redesigning the company around it. Over time the operations team shrank by roughly 97%.
Also inside: how AI gets convincingly wrong, where I keep a human in the loop, and the question I'd ask any owner:
Don't add AI to your company. Ask whether you would build the same company at all.
Full interview in my portfolio.
Where does your team still copy the same information between tools? Tell me the workflow here, or discuss it with Bosha on the site. We can start with one process and agree the scope before implementation.
bosha.ai
Stop hiring for operations. Redesign the company. — Valeriy Sterligov
An interview on rebuilding a business around AI: by his account, a roughly 97% smaller operations team, where AI fails, where people must stay, and the question every owner should ask.
The network for creativity
Join 1.25M professional creatives like you
Connect with clients, get discovered, and run your business 100% commission-free
Creatives on Contra have earned over $150M and we are just getting started
Related posts
Solea came to us in September at seed stage with a clear idea: scheduling is the core business problem, and AI is the right tool to solve it. They needed a design partner and within 24 hours of our first call, we completed a design test to prove we could move at their pace.
After launch, pest control businesses using the platform started running more appointments per day with the same headcount and reported a 10x ROI on the switch.
Solea signed their first national customer within 6 months, raised a $10M Series A, and we are continuing with them into the next phase.
Designed and built an AI operations dashboard that brings projects, decisions, tasks, research, connected tools, and voice interaction into one clear workspace. The goal is to turn complex workflows into a simple, usable system that helps people see what matters and act faster.
Seeing the voice interaction layer pull data from Supabase into the dashboard cuts down on manual context switching.
Building AI agents is easy. Building one you can trust with a real business workflow is harder.
I built a Freelance Admin Agent to explore that problem using Strands Agents SDK and Claude Sonnet 4.5 on Amazon Bedrock.
The working MVP takes manually provided client emails and turns them into a structured payment workflow:
→ Classifies invoice and payment emails
→ Extracts client, invoice, amount, currency, dates and contextual details using structured output
→ Stores and matches records in SQLite
→ Uses deterministic logic to identify full/short and on-time/late payments
→ Uses a second agent to draft a contextual follow-up only when needed
→ Keeps the final decision with the human through Approve / Edit / Reject
One of the most valuable parts wasn't adding another AI feature. It was finding a real reliability issue during testing: reprocessing the same email could duplicate records and inflate payment totals.
I reproduced it through repeat testing, traced the database behavior, and fixed it.
That reinforced something I'm increasingly applying to my AI engineering work:
A useful AI product isn't just an LLM call. It needs clear boundaries between AI reasoning, structured data, deterministic business logic, persistence, testing, and human control.
The next iteration will focus on stronger production safeguards, including source-grounded financial extraction, precise currency handling, stricter trust boundaries for untrusted input, and eventually live inbox integrations.
I'm interested in building more systems like this for startups and teams with repetitive workflows that can genuinely benefit from AI agents and automation.
Stack: Python · Strands Agents SDK · Amazon Bedrock · Claude Sonnet 4.5 · Pydantic · SQLite · Streamlit
turning client emails into admin tasks is such a real freelancer pain, nice build 🔥