VP Engineering

Reporting to the CEO

Where We Are Today:

We are two years into the development of a large-scale research analytics technology platform - work that was outsourced across a number of different vendors, not just in India but globally. That effort is now at its peak integration point, and the advent of AI has changed what done needs to mean before this product goes to market: it must first be reassessed and its IT infrastructure revamped, not simply finished as originally scoped.

Concretely, that means one of two paths - rebuilding the platform from scratch, or fixing the current architectural loopholes and integrating a number of disparate, vendor-built pieces into a single, seamless interface and experience for clients. Deciding which path is right, and then executing it, is the first major mandate of this role.

The Immediate Mandate:

We need a genuinely hands-on VP of Engineering - someone who will personally build the core transformation POD (a next-generation Solution/Technical Architect and a Senior DevOps, among others) and work directly with our CEO to revamp this platform and set the company's technology direction for at least the next five years.

Responsibilities:

Technical Strategy and Architecture:

- Define and publish a 3-year technology and AI architecture strategy with explicit sequencing, dependencies, and investment tradeoffs; revisit quarterly against evidence.

- Serve as final architectural authority on decisions with multi-year consequences: platform boundaries, data model, model strategy, build-versus-buy, and vendor concentration risk.

- Establish and personally staff a design review process that catches scale, cost, and safety failures before they reach production.

- Maintain a defensible technical moat: proprietary data assets, evaluation infrastructure, domain model performance, and workflow depth competitors cannot trivially replicate.

Applied AI and Machine Learning:

- Own the full model lifecycle: data curation, pretraining or continued pretraining where justified, post-training (SFT, preference optimization, distillation), quantization, serving, and deprecation.

- Lead architecture for agentic systems: planning, tool invocation, memory, multi-agent coordination, interoperability protocols, and failure containment when models behave unexpectedly.

- Build evaluation as a first-class engineering discipline - golden datasets, task-level benchmarks, human evaluation panels, automated judges, statistical significance in release gates, and continuous online measurement.

- Drive inference efficiency as a core competency: continuous batching, paged attention and KV-cache strategy, speculative decoding, model routing and cascading, distillation to smaller models, and hardware-aware serving.

- Set the standard for responsible AI: structured red-teaming, adversarial and prompt-injection testing, bias and fairness evaluation, model documentation, provenance and audit trails, and human oversight where consequences warrant it.

Platform, Infrastructure and Reliability:

- Own multi-region, cell-based architecture with defined blast radius, graceful degradation, load shedding, and tested failover.

- Set and enforce SLOs and error budgets; own incident command, blameless postmortems, and the closure of systemic root causes.

- Direct GPU and accelerator fleet strategy: capacity forecasting, topology-aware scheduling, utilization targets, reserved-versus-spot economics, and multi-cloud or hybrid positioning.

- Institutionalize platform engineering that raises developer productivity measurably - build and test infrastructure, environment parity, progressive delivery, and internal AI-native tooling.

Data:

- Own the data platform: streaming ingestion, lakehouse architecture, transformation, retrieval and vector infrastructure, feature and context stores.

- Enforce governance: classification, lineage, access control, retention, residency, deletion guarantees, and reproducibility of any model output under audit.

- Design and defend the data flywheel that compounds product quality release over release.

Security, Privacy and Compliance:

- Own security architecture and posture: zero-trust access, tenant isolation, secrets and key management, confidential computing where warranted, secure SDLC, and software and model supply-chain integrity.

- Lead certification and regulatory readiness relevant to our markets - SOC 2 Type II, ISO/IEC 27001, ISO/IEC 42001, HIPAA, PCI DSS, GDPR, EU AI Act, NIST AI RMF, DPDP - as a durable program rather than an audit sprint.

- Own the technical response to security incidents, disclosure obligations, and customer and regulator escalations.

Organization and Talent:

- Build a world-class engineering and science organization: define levels, calibration standards, promotion criteria, and a hiring bar that does not soften under delivery pressure.

- Recruit and develop senior leadership beneath you - VP Engineering, Head of AI, Head of Infrastructure, CISO or equivalent - and build genuine succession depth.

- Establish dual career ladders so principal and distinguished engineers can grow without leaving technical work.

- Set engineering culture: written technical communication, rigorous review, operational ownership, and intolerance for undocumented complexity.

Business and External Leadership:

- Partner with the CEO, CPO, and CFO on roadmap, pricing and packaging, capacity investment, and gross-margin trajectory.

- Represent technology in board and investor forums; lead technical diligence in fundraising and any acquisition activity.

- Engage directly with strategic customers, their security and architecture leadership, and standards or industry bodies where influence matters.

Minimum Qualifications:

- Bachelor's degree in Computer Science, Electrical Engineering, Mathematics, or a related technical field, or equivalent practical experience.

- 15+ years of software engineering experience, including 8+ years leading engineering or applied science organizations, with 3+ years managing managers.

- Experience leading an organization of 50+ engineers with responsibility for hiring, performance, and organizational design.

- Experience owning production systems at scale - high-throughput distributed services with formal availability commitments and 24x7 operational accountability.

- Demonstrated ownership of machine learning systems in production, including at least one large language model or generative AI product delivered to real users at scale, not as a pilot.

- Hands-on technical depth in at least one modern language (Python, Go, Java, C++, TypeScript, or Rust) sufficient to review architecture and code credibly.

- Experience with cloud infrastructure at scale (AWS, Azure, or GCP), container orchestration, and infrastructure as code.

- Experience owning security, privacy, or compliance obligations in an enterprise or regulated context.

- Experience presenting technical strategy and risk to executive leadership or a board.

Leadership Expectations:

- Depth without bottlenecking: You go deep where the risk is concentrated and delegate everywhere else, with clear decision rights.

- Written clarity: You reason in documents. Strategy, tradeoffs, and postmortems are legible to engineers, executives, and auditors alike.

- Judgment under ambiguity: You distinguish reversible from irreversible decisions and calibrate speed accordingly.

- Commercial fluency: You connect architecture to margin, sales cycle length, and retention without being asked to.

- Raises others: Your former reports are running organizations elsewhere.

- Calibrated on AI: Neither hype-driven nor dismissive. You know precisely what current systems do reliably, where they fail, and what to put between a model and a customer.

- Integrity under pressure: You do not ship what you would not stand behind in an incident review, a security audit, or a regulatory inquiry.

What Success Looks Like:

Horizon:

1. First 30 days:

- What's live: Full architecture assessment; roadmap and team plan agreed with the CEO.

- Why it matters: Sets the mandate in motion with executive backing from day one.

2. 3 months:

- Core backend re-architected onto scalable, AI-ready infrastructure.

- The single biggest technical bet the company is making right now.

3. 12 months:

- Next-gen product(s) built on the new infra generating measurable revenue.

- This is what the revenue-share component is tied to.

What You Inherit vs. What You'll Build:

- Inherit: three years of product and automation groundwork, a stable 600+ person company with organic profitable growth, and an executive team that has already committed to this transformation.

- Build: the entire architecture itself, the engineering team and hiring plan to support it, and the AI-native products this infrastructure will power.