Nijeesh NJ

AI Engineer — builds LLM products end to end

I'm an engineer who builds with a designer's eye, from model to interface. Final-year CS student ('27). Interned at Ankor, where I still ship to XAM, and at Tech Mahindra. I build with Python, TypeScript, Next.js, Rust and custom agents.

Full-timeBengaluruHire me View resume

01

Experience

Newest first. Tap a role for details.

AI + Full Stack Developer InternAnkor · XAM exam & proctoring platform · RemoteCompletedDec 25 – Mar 26

Core engineer on XAM, Ankor's AI-enabled exam and proctoring platform, and still contributing after the internship: 75+ merged PRs across the candidate app, API, admin console and billing.
Built the candidate exam flow end to end: session lifecycle, proctoring gates for fullscreen exit and window switching, candidate messaging, and a TipTap question editor.
Built admin.xam, a Next.js operator console for workspaces, plans, coupons and read-only content inspection.
Owned billing: subscriptions and one-off payments, coupon and first-month-free plans, GST invoicing, plan changes and idempotent checkout.
Migrated report generation to Rust, shipped transactional email, and built the ExAi assistant interface.
Led a reliability and security hardening pass across auth, sessions, evaluation, billing and workspace access, with CI checks for schema drift and migrations.

AI Engineer InternTech Mahindra · INDUS Project · BengaluruCompletedOct 25 – Dec 25

Built multilingual data collection and curation pipelines (OCR, data cleaning, QnA generation and a custom web crawler) for Hindi and low-resource Indian languages.
Processed and validated datasets such as FLORES-200, XNLI, XQuAD and Indic parallel corpora.
Analysed data for noise, translation inconsistencies and dataset bias.

AIML InternSkill First Labs · 3 production AI productsCompletedSep 24 – Aug 25

Job Market Analysis: an internal tool with a custom dataset and mapping algorithms inside an agentic workflow, automating internal research.
AI-powered question generation software for mass hiring.
Skill development software for students, with adaptive learning paths.

02

Achievements

A few hackathons and open source.

UI/UX Revamping CompetitionIIT PatnaWinner
Machine Learning HackathonNIT TrichyRunner-up
Generative AI Buildathon100x EngineersFinalist
AI Ascend HackathonAWSFinalist
GRASP HackathonASMEFinalist
Smart Transit ChallengeToyota Mobility FoundationContributor
Mesa, agent-based modelingOpen sourceContributor

03

Projects

Click any card to open it.

Trading Town

Open source

Agent-based artificial stock market — a deterministic, unit-tested Rust engine (Santa Fe ASM) where traders evolve forecasting rules via a genetic algorithm, wrapped in a Tauri 2 + PixiJS desktop world.

Case study GitHub

minimo — 8 sessions (active/idle) · 1 active · ~$369.62 API-equiv claude-code ✓ opencode ✓ codex ✗ (~/.codex not found) kimi-code ✗ STATUS PROVIDER MODEL CONTEXT LIFETIME COST RATE ● claude-code claude-sonnet-4-6 [████████░░] 14%/1M 742.5M ~$369.62 73K/m ○ claude-code claude-opus-4-8 [████░░░░░░] 8%/1M 40.1M ~$34.94 31K/m ○ opencode deepseek-v4-flash - 3.9M $0.0000 15K/m

Minimo

Open source

Observability TUI for AI coding agents — monitors Claude Code, OpenCode, Codex and Kimi sessions in one view, tracking context, cost and token burn in real time. Ships as a single Go binary.

Case study GitHub

Nite

Live

A Cursor for documents — an AI workspace with a two-pass (search + generation) agent pipeline, SSE streaming, and a three-layer content model (Markdown, blocks, TipTap).

Case study Live