I build AI systems that show their work.1
HireLens, my open-source ranking engine, ties every score to the exact resume text behind it. Agents, RAG pipelines, and the backends that ship them.
PostgreSQL at scale
- Owned PostgreSQL schema for a 200M-row ledger; cut slow queries 60%.
Selected work
HireLens
Scores a resume against a rubric you control. The model quotes its evidence, and code checks the quote is really in the resume before the score counts.
How HireLens scored a candidate
Event-driven & queue architecture
The candidate built event-driven services and designed idempotent webhook delivery with a Postgres-backed outbox table, which aligns with designing queues with retries or ordering. They also built retry queues with exponential backoff, demonstrating a solid understanding of queue architecture.
PostgreSQL at scale
The candidate owned a PostgreSQL schema for a 200M-row ledger and cut slow queries by 60%, indicating experience with tuning queries and designing schemas for large tables. This demonstrates solid experience with PostgreSQL at scale.
Distributed systems fundamentals
The candidate designed idempotent webhook delivery and built retry queues with exponential backoff, showing experience handling retries and idempotency in services. This aligns with the 'Basic' level of handling these aspects in one service.
Reliability & incident response
The candidate introduced SLO dashboards and on-call runbooks, leading to a 45% reduction in MTTR. This demonstrates driving incident response and improving MTTR, aligning with the 'Solid' level.
Fintech / payments context
The candidate led the migration of a payments platform and owned a PostgreSQL schema for a 200M-row ledger. This indicates direct work on payments or settlement services, fitting the 'Solid' description.
Kubernetes & infrastructure as code
The candidate led the migration of the payments platform to Kubernetes and lists Kubernetes and Terraform in their skills. This suggests they owned Kubernetes infrastructure for production, aligning with the 'Strong' level.
Jordan Avery Senior Backend Engineer — San Francisco, CA — jordan.avery@example.com EXPERIENCE Senior Backend Engineer — Acme Corp — 2021–present - Led migration of the payments platform to Kubernetes, cutting deploy time 80%. - Built event-driven services in TypeScript on Node.js handling 40k req/s. - Designed idempotent webhook delivery with a Postgres-backed outbox table. - Introduced SLO dashboards and on-call runbooks; MTTR down 45%. Backend Engineer — Northwind Systems — 2018–2021 - Owned PostgreSQL schema for a 200M-row ledger; cut slow queries 60%. - Built retry queues with exponential backoff for third-party integrations. - Participated in weekly incident reviews and postmortems. SKILLS TypeScript, Node.js, Go, PostgreSQL, Redis, Kafka, Kubernetes, Terraform, AWS EDUCATION B.S. Computer Science, State University (2014–2018)
■■■■■■■■■■■■ Senior Backend Engineer — San Francisco, CA — ■■■■■■■■■■■■■■■■■■■■■■■■ EXPERIENCE Senior Backend Engineer — Acme Corp — 2021–present - Led migration of the payments platform to Kubernetes, cutting deploy time 80%. - Built event-driven services in TypeScript on Node.js handling 40k req/s. - Designed idempotent webhook delivery with a Postgres-backed outbox table. - Introduced SLO dashboards and on-call runbooks; MTTR down 45%. Backend Engineer — Northwind Systems — 2018–2021 - Owned PostgreSQL schema for a 200M-row ledger; cut slow queries 60%. - Built retry queues with exponential backoff for third-party integrations. - Participated in weekly incident reviews and postmortems. SKILLS TypeScript, Node.js, Go, PostgreSQL, Redis, Kafka, Kubernetes, Terraform, AWS EDUCATION B.S. Computer Science, State University (2014–2018)
Real output from the HireLens engine (gemini-2.5-flash), fetched from the live demo on 15 Oct 2026. The candidate, company and job are fictional. Every quote is checked against its offsets when this site is built. SearchScribe AI Studio
A topic in, a researched draft out. The model picks source passages by ID and returns structured JSON; code owns the quotes, the HTML and the sanitising.
ForecastGPT
Quarterly results and earnings calls in, a structured outlook out. Code extracts the numbers; the model writes the narrative from inputs it is handed.
yDirect
A published Chrome extension for reusable snippets and prompts. Copying stays on the device; anything that touches the cloud is authorised by the server, not the UI.
Also built
- WriteRightSelf-hosted grammar checks and AI rewrites, reviewed before anything changes.
- REFLEX-AIA research agent that learns from its mistakes and won’t answer without a citation.
- News-to-Video GeneratorHeadlines and images in, narrated MP4 news videos out.
- CricketStat ProA T20 World Cup analytics pipeline feeding a Power BI selection dashboard.
- AI-Powered 3-Factor LoginFace, then email OTP, then a voice phrase, with liveness checks against spoofing.
Experience
Translytics
AI and backend engineering on a multi-tenant supply-chain planning SaaS.
- Ship LLM-backed features and Python backend services, from design through QA and release.
- Design REST APIs and background jobs for data-heavy planning workflows, and improve how data moves in and out of the platform: imports, validation and exports.
Tata Elxsi
Two years of production Python, then a deliberate move into generative AI.
- Built a tool that turns legacy test cases into Gherkin feature files, with a web UI to review and export, cutting the manual hand-off between QA and development.
- Built FastAPI apps on LLMs and RAG (LangChain, LlamaIndex, Chroma) in the Generative AI bootcamp.
- Wrote Python that turns OBD diagnostic data and trouble codes into reports with automatic OK / Not-OK status.
- Validated ECU configurations over UDS and traced faults to root cause with Center of Excellence engineers.
PHN Technology
An end-to-end sports analytics pipeline for the T20 World Cup.
- Scraped ESPN Cricinfo, cleaned it in pandas and modelled a star schema in Power BI.
- Shipped a selection dashboard with role filters, drill-downs and head-to-head comparison.
BitMap Technology
A three-factor login for online banking: face, then email OTP, then a voice phrase.
- OpenCV detection and an LBPH recogniser, with a calibration tool for thresholds.
- In testing: about 95% correct identification, and 88% of spoof attempts blocked.
Capabilities
- LLM applications
Schema-validated structured output, behind a provider layer that swaps models.
- RAG and retrieval
Chunking, embeddings and vector search, tied back to source passages.
- Agents and evaluation
Agents that cite or refuse, and evals that measure bias instead of assuming it.
- Backend and data
Typed APIs, job queues, and auth with rotating refresh tokens.
- Shipping
Next.js, Chrome extensions, Docker and CI, released to npm and the Web Store.
About
I like building things that actually work.
Tools, workflows, AI systems, and small utilities that make someone’s day easier. I start from the root of a problem, simplify it, and shape it into something structured and dependable.
Most of what I build now follows one rule: the model proposes, and code checks. HireLens won’t count a score whose quote isn’t in the resume. SearchScribe never lets a model write HTML.
Good engineering is clarity, not complexity: predictable behaviour, and decisions I can explain.
At college I coordinated Team Pragyaa, one of the institute’s largest national socio-technical fests, and volunteered at Smart India Hackathon 2020.
Education & credentials
B.Tech, Computer Science
Shri Guru Gobind Singhji Institute of Engineering & Technology
AI Fluency: Framework & Foundations
OCI 2025 Certified AI Foundations Associate
Contact
Open to AI engineering roles and interesting problems.
The fastest way to reach me is email. I read everything and reply to anything specific.