Head - Engineering

Head of Engineering

About AdvaRisk:

AdvaRisk is building the property risk intelligence layer for Indian lending. Banks, NBFCs, and HFCs spend enormous manual effort verifying property collateral reading title documents, encumbrance certificates, and registry records across dozens of states, languages, and formats to make lending decisions under RBI and NHB frameworks. We replace that manual, error-prone work with an API-first platform: ML that performs OCR, NER, transliteration, and RAG powered LLM based address matching across Indian languages, resolves property identity through a DAG-based lineage model (UPID), and delivers collateral risk assessments in real time.

We process a high volume of property evaluations daily for top 3 private banks, NBFCs, and HFCs. Our investors include BIG Capital, NAB Ventures (venture investing arm of NABARD), ICICI Bank & SEA fund. We are at the point where the platform has to go from fast-moving to genuinely production-grade at scale and where every API response a lender builds a credit decision on has to be right.

The Role:

We have live integrations with regulated lenders and multiple AI-driven products in production. We're looking for a Head of Engineering to own how engineering plans, builds, ships, and operates.

You will run all of product and platform engineering the application teams across our ingestion, property-intelligence, risk-scoring, and API/delivery streams; the platform, infrastructure, and reliability function; DevSecOps; and frontend. The engineering leads across these areas will report to you. You report to the Co-founder leading Product & Technology and take day-to-day ownership of engineering execution.

This is not a hands-off role.

You'll set technical direction, sit in design reviews, get into production incidents, and solve hard problems alongside strong senior engineers while building the process and standards that let a ~30-person team ship like a much larger one.

What You'll Own:

- All engineering execution. Application engineering across the value chain data acquisition, property intelligence, risk scoring, and the customer-facing API plus platform/infrastructure, reliability, and frontend. You own what ships, when it ships, and how good it is.

- The API contract and production reliability. Our customers are regulated lenders making credit decisions on our output. A wrong collateral assessment or a broken API contract isn't a bug it's a lending risk and a trust event. Own observability across the pipeline, catching data-quality regressions and pipeline failures before customers do; API versioning and backward compatibility discipline (banks integrate against your contract and cannot absorb breaking changes); incident response; and a real testing and code-quality culture.

- The Property Intelligence core. The DAG-based lineage model, UPID resolution, and deduplication engine are the company's moat. Protect the stability and coherence of this subsystem above everything else, and staff it with your most senior, most stable engineers.

- Ingestion resilience. Data sources are adversarial and brittle. Own the data ingestion reliability, source diversity, and the discipline that ensures one broken registry never halts evaluations.

- Platform, infrastructure, and DevSecOps. The Python, Django, Fastapi, PostgreSQL stack; deployment and observability. Security and compliance engineering is core, not bolted on: data residency (data stays in India), PII handling on property- and borrower-linked data, tamper-evident audit trails, and readiness for SOC 2 / ISO 27001 and RBI IT-governance scrutiny.

- Customer go-lives. Onboarding a regulated lender is high-stakes across hosted and on-premise/single-tenant setups, each with its own security review, VAPT, and compliance requirements. Make deployment to new customer environments fast and repeatable.

- Customer-facing engineering and communication. You are a technical face of AdvaRisk to our customers. Engage directly with lenders' engineering, IT, security, and compliance teams in integration discussions, architecture and security reviews, escalations, and incident communication. Explain what happened, why, and what we're doing about it in language a bank's technical and risk stakeholders trust. Every one of these interactions either builds or erodes the credibility our product depends on; own that, and set the standard for how your team shows up in front of customers.

- The engineering operating system. Planning and execution cadence, design and code review, release process, on-call, and the bar for what done means. Replace heroics and tribal knowledge including our state-by-state registry idiosyncrasies with systems and documentation.

- AI-first engineering. Make AI a default part of how the team builds, not a side experiment. Drive adoption of AI-assisted coding, review, testing, and documentation to raise the leverage of every engineer; embed AI and agentic tooling into the engineering workflow and internal ops; and set the standards for using it responsibly in a regulated environment respecting data residency, PII handling, and our security posture, so no customer or borrower data leaks into third-party tools. You'll set the bar for what good AI-assisted engineering looks like and hold it.

- The team and the bar. Set leveling and promotion standards, run performance honestly, and hire across the org as we scale. Define the structure as engineering grows from a handful of pods into a real organization.

- Partnership with Data Science and Product. Work with the Data Science team on model serving, evaluation, accuracy benchmarks, and the AI-MLOps seam establishing shared model contracts and production-metric ownership so models don't drift silently. Work with Product to turn the UPID strategy into reliable, shippable software.

- Technical strategy. Think beyond the current sprint. Shape the multi-quarter technical direction that turns the UPID vision into architecture and roadmap; weigh build-vs-buy, sequencing, and technical debt against business priorities and our regulatory reality; and connect engineering decisions to what the business is trying to achieve. You'll be a partner in company and product strategy, not just an executor of it bringing an engineering-grounded point of view to the table and translating it into a plan the team can build.

What Success Looks Like in the First 90 Days:

- Days 130: Learn the systems, the products, the team. Read the architecture and the code. Use every product. Look at raw AI-ML extraction output OCR, NER, LLM based matching against source registry documents and understand where the platform is fragile and why. Sit in on a lender deployment and a production incident. Meet every engineer, every lead, and DS and Product leadership. By the end of month one, you can name the three things most hurting reliability and delivery speed, and give a candid read on the strengths and gaps of the team and its leads.

- Days 3060: Take ownership and set the operating system. Take over day-to-day running of engineering planning, reviews, prioritization, delivery accountability. Stand up the operating cadence: how teams plan and commit, how code and designs get reviewed, how releases ship, how incidents are handled. Pick the single biggest reliability or delivery problem API stability, ingestion resilience, or model-serving quality and personally drive it to a fix, so the team sees the new bar rather than just hearing about it.

- Days 6090: Show the bar moving and stand up the plan. Demonstrate measurable improvement on reliability and release quality, and a faster, more standardized path to customer go-live. Have a hiring and structure plan for the next two quarters.

What You'll Bring:

- 12+ years building production software, with 4+ years managing engineers including managing leads or managers, not just individual contributors.

- A track record of running engineering for a multi-team or multi-product B2B SaaS company ideally data- or API-intensive and taking products from fragile to genuinely production-grade at scale.

- Real technical depth. You can reason about architecture, review a design, and debug a production issue. Strong senior engineers respect you because you understand the system, not just the org chart.

- Hands-on experience with cloud infrastructure, containers/Kubernetes, CI/CD, event-driven or queue-based systems (RabbitMQ/Celery or equivalents), PostgreSQL at scale, and data- or document-heavy backend systems with the judgment to know where reliability actually breaks.

- Built engineering process and quality culture from a low baseline in a fast-moving environment testing, observability, release discipline, incident response, API versioning without grinding delivery to a halt.

- A record of hiring, leveling, and developing engineers and managers, and of setting and holding a high bar for shipped software.

- Clear, direct communication. You make hard calls with incomplete information, push back when requirements are wrong, and don't wait for permission. High agency, comfortable with ambiguity.

- Willing and able to work in-office in Pune.

Bonus Points If:

- You've worked in fintech or another regulated domain particularly with banks, NBFCs, or HFCs, and RBI/NHB-style regulatory and audit requirements.

- You've run engineering for an API-first infrastructure product where AI-ML accuracy is the core product value, not a feature bolted on including model serving, evaluation pipelines, and AI-MLOps.

- You've shipped and supported enterprise deployments in customer-controlled cloud or on-prem/private environments with serious security and compliance requirements.

- You've scaled an engineering organization through hypergrowth and lived through the structural and cultural breakpoints.

- You have hands-on familiarity with our stack Python, Celery, RabbitMQ, PostgreSQL, Angular, multi-cloud.

- You've worked with multilingual/Indian-language data, OCR/document AI, geospatial data, or land-records / property-title domains.

Why Join Us:

- Category-defining infrastructure. You'll build the property risk and identity layer the UPID vision that Indian lending is missing. This is prerequisite infrastructure, not a point tool.

- Hard, data-intensive problems. ML, regulated fintech, and scale intersect here: adversarial ingestion, entity resolution across states and languages, and an API that regulated lenders bet credit decisions on.

- Real-world impact. The systems your teams build make property-backed lending faster, safer, and more accessible across India.

- World-class team and ownership culture. Work alongside strong talent across engineering, data science, and product, in a fast-paced, ownership-driven environment.

Skills

CTO, Technical Architect, IT Jobs in BFSI, IT Product Development