Senior AI Forward Deployed Engineer - Cogentiq I2C

Fractal; Bengaluru; On-site; INR 2,500,000 - 4,500,000; Full time

Job summary

Fractal seeks a Senior AI Forward Deployed Engineer to own technical outcomes for a portfolio of strategic accounts, turning business problems into production AI solutions in collaboration with finance teams and IT leadership.

You will lead deployment initiatives, define KPI contracts, and own integration with ERP and document infrastructure while shaping product roadmaps through reusable assets and governance guidance.

Qualifications

• 6 to 10 years building and operating production systems, including at least two years in customer-facing delivery.
• Strong Python and system design depth.
• Depth across LLM and agentic architectures, retrieval, evaluation methodology, guardrails and observability.
• Cloud deployment experience at scale; Azure is preferred; containers and orchestration.

Responsibilities

• Own the technical outcome for a portfolio of accounts from discovery through production rollout and scale.
• Set and defend the deployment KPI contract: accuracy, throughput, latency, cost per transaction, adoption and business impact.
• Lead a deployment team of Forward Deployed Engineers and specialists; scope, sequence and unblock work.
• Act as incident commander for production failures; perform root cause analysis and remediation.
• Explain system capabilities and residual risks to finance and IT leadership; guide governance and responsible AI.

Job description

It's fun to work in a company where people truly BELIEVE in what they are doing!

We're committed to bringing passion and customer focus to the business.

Senior AI Forward Deployed Engineer - Cogentiq I2C (Invoice-to-Cash)

Role Summary

You own the technical success of Cogentiq I2C at a portfolio of strategic accounts. You take a business problem that arrives stated in finance terms, reduce our days sales outstanding, stop writing off small-balance deductions, and turn it into a production AI system that the customer’s finance organisation actually uses.

You lead a small deployment team. You make the architecture calls. You are the person the customer sponsor calls when something is wrong, and the person who tells them what we will not build. This is an engineering role. You will still be in the codebase. The difference from the first band is that you also carry the account, the design and the team.

What You Will Own

• Own the technical outcome for a portfolio of accounts, from discovery through production rollout and scale.
• Set and defend the deployment KPI contract: accuracy, straight-through processing rate, latency, cost per transaction, adoption and business impact.
• Hold the technical relationship with customer sponsors and IT leadership.

Solution design and architecture

• Translate ambiguous goals into a solution design covering agent topology, retrieval strategy, human oversight boundaries, integration approach and data model.
• Make and document the build versus configure versus extend-the-product calls. Decline bespoke work that cannot be maintained across the customer base.
• Own the integration design against customer ERP, banking and document infrastructure.

Delivery leadership

• Lead a deployment team of Forward Deployed Engineers and specialists. Scope the work, sequence it and unblock it.
• Own proof-of-value and pilot scoping, including effort estimates and success criteria that can be measured rather than asserted.
• Act as incident commander for production failures. Own the root cause analysis and the remediation plan.

Advisory

• Explain to finance and IT leadership what the system can and cannot do and what residual risk remains after it is deployed.
• Guide customers on governance, human oversight design, model risk and responsible AI.

Product and asset leadership

• Convert deployment learnings into reusable assets: reference architectures, connector templates, evaluation packs and deployment accelerators.
• Bring evidenced feedback to the product roadmap. Distinguish a single customer’s preference from a genuine product gap.

What You Bring

• 6 to 10 years building and operating production systems, including at least two years in customer-facing delivery.
• A track record of taking at least one non-trivial AI system to production and keeping it running afterwards.
• Strong Python and system design depth.
• Depth across LLM and agentic architectures, retrieval, evaluation methodology, guardrails and observability.
• Cloud deployment experience at scale. Azure is preferred given our stack. Containers and orchestration.
• Enough finance vocabulary to hold a conversation about DSO, deductions, remittance advice, and cash application without a translator or the evident ability to get there within a quarter.
• Commercial judgement. You can recognise when a customer request has quietly become unpaid scope, and you raise it early.
• Willingness to travel and embed with customers.

Nice to have

• AI deployments in regulated industries.
• Enterprise ERP integration experience, particularly SAP S/4HANA or Dynamics 365 Finance and Operations.
• Experience mentoring or leading engineers.
• Pre-sales or solution architecture exposure.
• Startup or product-building experience where customer feedback shaped what got built.

What Success Looks Like In Your First Year

• Two accounts running in production against agreed KPIs.
• The Forward Deployed Engineers on your team are measurably more capable than when they joined it.
• At least two reusable assets you produced are in use by another deployment team.

If you like wild growth and working with happy, enthusiastic over-achievers, you’ll enjoy your career with us!