What are you building?
I take freelance work in AI product engineering, solo end-to-end platform delivery, and AI-assisted development enablement. Fifteen years in distributed systems and cloud-native backends; three production platforms shipped solo. Remote, from Noida, working with teams in most time zones.
Employed full-time; taking one freelance engagement at a time. That is a real constraint rather than scarcity marketing — it means the work gets done properly, and it means timelines depend on what is already running.
Tell me the problem rather than the solution you have in mind — it is usually a shorter message and a better one. If it is not something I should take on, I will say so and point you somewhere better.
- Emailmahajananuj07@gmail.com
Fastest route to a reply. A short description of the problem is enough to start.
- LinkedInanuj-mahajana2063790
Career history, and fine for an introduction.
- GitHubmahAnuj
The code, including everything referenced on this site.
What an engagement usually looks like
AI product engineering
LLM pipelines, retrieval-augmented generation, agent tooling, MCP servers and evaluation harnesses — built to run in production rather than to demo. Typically a defined build with a working system at the end.
4–12 weeks · fixed scope or retainer
Solo end-to-end delivery
A complete production platform delivered by one engineer: architecture, data pipelines, infrastructure, deployment, and admin tooling your team can run without a developer. For products that need to exist before a team can be hired to maintain them.
8–20 weeks · fixed scope
AI-assisted development enablement
For teams already shipping who want real leverage from AI-assisted development without the review burden eating the gains — where to put guardrails, what to automate, what to keep human.
Workshop or ongoing retainer
What helps in a first message
- What the system has to do, and who it is for.
- What already exists — a codebase, a spec, a spreadsheet, nothing.
- When it needs to be live, and what happens if it is not.
- A budget range. It helps more than it hurts: it is the fastest way to find out whether we are talking about the same size of project.
If you would rather see the work first, the case studies cover all three platforms in detail, and the writing is mostly about what went wrong on them.