Open to Forward Deployed, backend & applied-AI engineering roles

Software Engineer · Backend & Applied AI

Shivank Bhardwaj

Systems that hold up.

I build backend systems and production AI agents — and I work directly with the people who use them, turning their needs into API contracts and shipped software. Currently at Bidgely, owning solar billing & time-of-use rate optimization end to end and building internal AI agents gated by evals.

Start here

Three ways in

New here? This site is organised around three things — what I've built, how I think, and who I am. Go wherever your question points.

01

Work

The problems I've owned end to end — AI agents with evals, billing systems, enterprise integrations, data migrations — written up as case studies with the decisions, tradeoffs, and failures left visible.

Start here if you want proof.

Browse case studies →
02

Thinking

Short essays on how I approach the craft: when to make a call fast versus treat it like a bet, how to debug by reasoning instead of guessing, and other lessons from real work.

Start here to see how I think.

Read the essays →
03

About

The short version of me — four years across aviation, finance, and energy, the stack I work in day to day, and what I’m looking for next.

Start here for the background.

Read my story →

Selected work

Case studies, not screenshots

bet01

Internal AI Agents for Backend Operations — Gated by Evals

Built internal AI agents that automate log analysis, anomaly detection, data validation, and incident summarization — with eval suites that catch regressions before every release.

PythonLLM APIsTool CallingRAG
Read case study →
architecture02

Owning Solar Billing & Time-of-Use Rate Optimization End to End

Own solar billing and time-of-use rate optimization end to end — API design, data modelling, release — turning seasonal and mid-cycle rate logic into accurate, repeatable costs.

JavaPythonREST APIsCassandra
Read case study →
alignment03

A Talent-Pool Tracking App, From Scratch to Production in Two Months

Took a talent-pool tracking app from scratch to production in 2 months — React SPA over a Node.js REST API, secured with Azure AD — by turning stakeholder needs into UI flows and API contracts.

TypeScriptReactNode.jsREST APIs
Read case study →
See all case studies →

Writing

From the archive

Evals Before Features: How I Ship AI Agents People Trust

An agent that works in a demo and an agent a team relies on are separated by one thing: knowing when a change made it worse.

→

Debugging as Detective Work: Stop Guessing, Start Reasoning

Most debugging wastes time by skipping the hypothesis step. Here's the frame that fixed that for me.

→
Read all essays →

Let's build systems that hold up.

Open to Forward Deployed, backend & applied-AI roles. Message me directly — by email or on LinkedIn, whichever you prefer.

theshivankbhardwaj@gmail.comMessage on LinkedIn
Résumé (PDF)GitHubFor recruiters