› ai can absolutely save your business serious time and money — when it's put on the right work, with the right guardrails.
› i'm the founder of alt software solutions, where i build small, reliable agents and automations for the work that eats your team's week: customer replies, quotes, reports, lead triage, document handling, internal q&a.
› best fit: businesses already running, where one specific process is costing more hours, more money or more customers than it should.
now.
- building ai agents that take repetitive work off your team — without growing headcount.
- helping small teams handle 3–5× more volume on the same work that used to need new hires.
- replacing scattered ai usage (personal chatgpt tabs, copy-paste prompts, no measurement) with a small set of agents your business actually owns.
before.
- led engineering teams from idea to production — owning architecture, delivery and team direction.
- shipped complex backend systems for an insurance platform and a global reinsurance group, where reliability isn't optional.
- worked hand-in-hand with product, sales, marketing and operations — and that cross-team reality is exactly where ai rollouts succeed or fail.
where i help.
your team spends hours every week on the same repetitive work — replies, quotes, reports, data entry — and you know there has to be a smarter way.
leads, tickets or orders are arriving faster than your team can handle, and the quality of follow-up is starting to slip.
information lives in five different tools and nobody has time to connect them, so decisions are slow and your numbers are always a week behind.
what this looks like.
support inbox triage — customer emails come in; an agent reads each one, drafts a reply in your voice, attaches the right help link, and only escalates the cases that actually need a human. small teams handle 3–4× more without hiring.
quotes & proposals — a salesperson forwards an inquiry; an agent extracts the scope, applies your pricing rules, drafts the proposal in your template, and the rep just reviews and sends. turnaround drops from days to hours.
weekly business pulse — every monday morning, an agent pulls numbers from your tools (quickbooks, stripe, hubspot, your spreadsheet), writes a one-page summary of what changed and what to watch, lands in your inbox before coffee.
document intake — pdfs, contracts, receipts, invoices land in a folder or inbox; an agent pulls the fields you care about, validates them, and pushes to your system. accounting spends time on edge cases, not data entry.
these are patterns, not products — every build is tailored to how your business actually runs.
engagements.
- opportunity audit — i look at how your business runs and tell you the top 2–3 places ai will save the most time, money or headcount cost.
- build & ship one — we pick the highest-leverage workflow and i build, test and deploy a working agent end-to-end, integrated with the tools you already use.
- ai stack cleanup — replace personal chatgpt tabs and one-off scripts with a small, reliable set of agents your team can actually own and trust.
what production means.
- clear success criteria before we build — what saves time, what saves money, and by how much.
- agents you can see working: every action logged, every failure visible, no black boxes.
- humans stay in the loop for anything touching money, customers or compliance.
- handed off clean — owned by your team, with docs a non-engineer can follow.
work together.
there's probably a process in your business that costs too much time, too much money, or too much of your team's patience. send me a short description and i'll tell you honestly whether ai is the right tool — and if it is, what i'd build first.
request a 15-min call →background.
- nearly a decade building production software across backend systems, apis, architecture and delivery.
- comfortable between product, engineering, qa and business operations — the place ai rollouts usually succeed or fail.
- university of belgrade, electrical engineering
- mathematical grammar school, belgrade