Open to 2027 grad roles ✦

Graduate Cloud & Software Engineer
Winchester, UK

Aman Kumar Sah

I build cloud, platform and AI products that stay useful after the demo.

01 About

Hi, I'm Aman (Kris to friends).

I'm a Computer Science (AI) student at Brunel University London, graduating June 2027, and a graduate engineer who cares about the unglamorous part of software: the bit that has to keep working after launch.

I worked as a placement software engineer at Faith Associates (Jun 2025 – Sep 2026), shipping React and TypeScript on live client platforms and recovering a compromised production server. I'm AWS Solutions Architect and Terraform certified.

Outside of work I turn hackathon ideas into working products, with wins at Google's Gemini Live Agent Challenge and NVIDIA Hack for Impact, plus a Brunel venture award for BrunelOnChain. I'm now looking for UK graduate software engineering, cloud, platform and AI product roles.

02 Experience

Where I've done the work

Jun 2025 – Sep 2026
Software Engineer (Placement) · Faith Associates
Worked on live client platforms where reliability, release quality and communication mattered.
- Built React and TypeScript features with REST API integration and release testing.
- Diagnosed production defects, supported incident response and documented confirmed causes.
- Handled platform recovery, server hardening, Cloudflare protections, backups and PM2 deployment.
Left a compromised live platform secured, monitored and documented - with a verified 15-minute recovery path.
Jan 2024 – Jun 2027
BSc Computer Science - Artificial Intelligence · Brunel University London
Graduating in June 2027, with focus across software development, algorithms, networks, operating systems and usability.
- Artificial Intelligence pathway with practical software engineering grounding.
- Applying course learning directly in hackathons, projects and placement work.
Graduating June 2027, with coursework applied straight into placement work and award-winning hackathon builds.
Sep 2022 – Nov 2023
Research & Development Analyst · Herald Development Platform
Explored AI-agent architectures and evaluated how retrieval, memory and ranking affected usefulness.
- Researched agent systems combining retrieval, memory, ranking and tool use.
- Compared retrieval quality, latency, false positives and memory persistence across revisions.
Gave the team evidence, not opinions, on which retrieval and memory designs actually made agents more useful.
Professional development
Cloud and AI development · Certifications & Training
Formal cloud training alongside hands-on delivery work and modern AI tooling practice.
- AWS Certified Solutions Architect Associate and Terraform Associate.
- AZ-104 Azure Administrator training and Hugging Face AI Agents Course completed.
Certified AWS Solutions Architect Associate and HashiCorp Terraform Associate.

Full experience timeline

03 Selected work

Problems I picked up, and what changed.

Cloud & Platform · 2026Builder · Microsoft AI Dev Days 2026

MigrationOps Copilot

Website migrations routinely ship hidden outages. DNS cutovers propagate unevenly, SSL certificates get forgotten on new load balancers and unmapped paths return silent 404s, and most teams still catch this with ad-hoc curl scripts.

Captured live SSL trust, DNS resolution and HTTP status and latency from source and target endpoints, then diffed them into a structured JSON report.
Chained bounded LLM stages (triager, diagnostician, planner) on Microsoft Agent Framework and Azure OpenAI, each reasoning only over captured data.
Gated a non-destructive, simulated executor behind explicit human approval, exposed through a CLI, a FastAPI web UI and an MCP server, with GitHub Actions CI.

A broken Azure App Service cutover is flagged critical, traced to a missing SSL binding and paired with a reviewable fix plan before users ever hit it.

PythonFastAPIAzure OpenAIMicrosoft Agent Framework+2

GitHub ↗

AI Systems · 2026Builder · Qwen Cloud AI Hackathon, Track 4

Signal

Growth teams keep having to decide which ad creative to refresh, retire or scale, with the evidence scattered across ad libraries and campaign pages. A generic chatbot gives them no workflow, no structured evidence and no accountability.

Collected live owned and competitor creative observations through Qwen web search and normalised them into typed creative families.
Added a second, adversarial Qwen pass that grades every factual claim and fetches every cited URL to check it against the live web.
Routed each run into an append-only review ledger where an operator must acknowledge every checkpoint, then deployed to Alibaba Cloud Function Compute.

Live endpoints return pre-audited decision packets held at pending human review: a decision with an audit trail, not a chat reply.

Qwen Model StudioAlibaba Cloud Function ComputeTypeScriptPython

GitHub ↗Live demo ↗

Platform recovery and infrastructure hardening work

Cloud & Platform · 2025Lead recovery engineer · Faith Associates

Platform Recovery & Hardening

A live client platform for British Muslim Trust had been compromised and was running unstable: no hardened edge, no automated off-site backups and no process supervision, so every traffic spike risked downtime.

Locked the origin to Cloudflare's IP ranges at the UFW layer, behind WAF rules and Authenticated Origin Pulls.
Replaced unmanaged Node processes with a PM2 cluster that has memory-bound restarts, zero-downtime reloads and log rotation.
Scripted daily database and file snapshots to off-site storage, then wrote the disaster-recovery runbooks and handover.

No unplanned outages after the rollout, direct-to-host scans shut out, and a verified 15-minute recovery time objective.

Cloudflare WAFPM2UbuntuUFW+3

AI Systems · 2026Builder · OpenAI Build Week

OneFirm

Small businesses facing a big call, like whether a 14-person agency can safely hire two strategists, need the combined work of a finance partner, strategist, auditor and risk adviser. Another dashboard or a generic chat doesn't give them that.

Built a deterministic 12-month cash model comparing current-team, one-hire and two-hire scenarios under growth, collection-timing and client-renewal stress.
Enforced explicit decision gates (£150k cash floor, £108k MRR, client concentration under 32%) and exposed the full monthly calculation trace.
Wrote the partner memo and dissent with the OpenAI Responses API and Zod structured outputs, instructed never to invent or recalculate a number.

One consequential question answered end to end: every figure is traceable, and every assumption can be challenged from the Finance, Strategy, Risk, Audit and Capital views.

TypeScriptReactViteNode.js+3

AI Systems · 2026Builder · Hack-Nation 2026 (Spiral challenge)

Proof402

As AI agents get wallets and call paid APIs, the hard problem isn't payment but permission. An agent that can pay can also burn a budget on a scam API or a fake grant.

Designed a policy gate that intercepts every risky agent action and answers with a real HTTP 402 Payment Required, ready for L402 clients.
Issued machine-readable proof receipts (verdict, confidence, evidence, risk flags) for 3–6 sat Lightning-style micropayments behind a swappable payment adapter.
Built a configurable spend policy (caps, confidence thresholds, blocked terms) and a live dashboard of allowed, denied and manual-review actions.

Every agent spend leaves an auditable reason (allowed, denied or sent to manual review), using a 402 and receipt schema ready to plug into real Lightning rails.

Next.js 15TypeScriptTailwind CSSHTTP 402 / L402+1

GitHub ↗

Product & Native · 2026Builder · Encode Vibe Coding Hackathon

RUUM

Second-hand furniture listings rely on flat photos and guessed dimensions. Buyers can't tell whether a piece will actually fit, so good furniture never gets reused.

Built a native SwiftUI scan flow on Apple Object Capture that reconstructs a USDZ model on-device, with bounds-derived dimensions and a SHA-256 model hash.
Shipped a Vercel TypeScript backend for AI listing generation, buyer interest and asset storage in Vercel Blob.
Minted a Walrus-backed Sui reuse passport for each item, with the whole system specified in codeplain specs.

A working scan-to-listing loop with real on-device 3D capture, real dimensions for fit-checking and a verifiable reuse record.

SwiftSwiftUIObject CaptureTypeScript+3

See all work

04 Skills

Skills, with receipts.

Cloud & Ops

AWSTerraformAzureGoogle Cloud RunGKECloudflareDockerCI/CDLinuxPM2

Hardened a compromised live charity platform behind Cloudflare, a PM2 cluster and off-site backups, and shipped sandboxed agent desktops on Cloud Run and GKE.

Proof:Platform Recovery & Hardening SwarmOS Cloud Backend MigrationOps Copilot

AI systems

Agent workflowsRAGTool callingStructured outputsEvaluationHuman-in-the-loop

Built agents that check their own work: adversarial fact-checking, permission-aware retrieval and approval gates before anything risky happens.

Proof:Signal TrustRAG Proof402

Backend & APIs

PythonFastAPINode.jsREST APIsAuth & RBACPostgreSQL

Designed deny-by-default auth, deterministic financial models and HTTP 402 policy gates that leave an audit trail behind every decision.

Proof:Gatekeeper OneFirm Proof402

Frontend

TypeScriptReactNext.jsTailwind CSSVite

Shipped React and TypeScript features with API integration and release testing on live client platforms during placement.

Proof:OneFirm BrunelOnChain

Native & mobile

SwiftSwiftUIObject CaptureKotlinJava

Built an iOS scan-to-listing app that reconstructs true-to-size 3D furniture models on-device.

Proof:RUUM

05 Proof

Wins under a deadline

Global UI Navigators Winner

Google Gemini Live Agent Challenge

Winner

NVIDIA Hack for Impact London

Venture competition award

Brunel venture competition - BrunelOnChain