A ranked backlog
Every candidate scored on value, data readiness and risk, so the funding argument is already made before a sprint is booked.
AI Core — Where we lead
AI Consulting
A costed roadmap that says which AI ideas to fund, which to shelve, and what each one is worth.
- AI readiness & data audit
- Use-case scoring and ROI modelling
- Build-vs-buy and model selection
- Governance, risk and compliance guardrails
- Engagement shape
- Fixed-fee discovery sprint run with two or three of your operators and one executive sponsor.
- Typical duration
- One to two weeks of fieldwork, plus a readout week.
- Starting point
- A workshop covering the departments where the manual load is heaviest.
Overview
What the work involves.
Most AI programmes stall because the first project was chosen in a meeting rather than measured against the work. We start with the work. We sit with the people who clear the queue, watch what they actually do, and count the hours it takes them.
Every candidate use case is then scored on three axes: the value it releases, the readiness of the data behind it, and the risk it carries when the model is wrong. Some ideas survive that. Most do not, and saying so in week two is the point of the exercise.
You finish with a ranked backlog, a cost model per use case that includes inference and maintenance rather than build alone, and a build-versus-buy call on each. Where the honest answer is that a rules engine beats a model, the report says that.
Outcomes
What you get.
A cost model you can defend
Engineering, inference, monitoring and the human review that stays in the loop, priced per workflow instead of as one programme number.
Build-versus-buy clarity
Where a product on the market wins, we say so. Where it cannot reach your data or locks you in, we show what building instead costs over three years.
Guardrails agreed early
Data handling, model access, approval gates and audit expectations written down while they are still cheap to change.
Deliverables
What lands in your hands.
- Workflow map of every department in scope
- Use-case scorecard rating value, data readiness and risk
- Data readiness report per system of record
- Cost model covering build, inference and running cost
- Build-versus-buy recommendation for each candidate
- Twelve-month roadmap with sequencing and dependencies
- Governance and risk register
Approach
How we run it.
Sit with the operators
Two days with the teams doing the work. We record how a task actually moves, where it queues, what people do when the system says no, and which spreadsheet holds the truth.
Audit the data
Each candidate is traced back to the systems that hold its data. We test whether we can reach it, and note what is missing, stale, duplicated or typed in by hand every morning.
Score and cost
Value comes from observed volumes and handling times, not from a vendor benchmark. Risk gets a class. Inference is priced at the token volumes your real traffic would produce.
Write the roadmap
Sequenced so the first build funds the second. We present it to your leadership, take the argument, and revise until the plan survives the room.
Each phase ends with something you can read and act on. If the evidence says stop, stopping there is a supported outcome rather than an awkward conversation.
Stack
What we build it with.
- Anthropic Claude
- OpenAI
- Azure OpenAI Service
- Amazon Bedrock
- Snowflake
- dbt
- LangSmith
- Metabase
Tooling is a decision we make per project, against your constraints and whatever your team already runs. Nothing on this list is a requirement, and we will work inside your existing stack where it holds up.
Questions
What clients ask about AI Consulting.
Often the recommendation is to buy, or to fix a process before automating it. The roadmap is written to stand alone — clients have taken it to a different delivery partner and it still works.
The scorecard and the cost model land at the end of the sprint, one to two weeks in. Nothing is held back for a paid second phase.
Yes. A focused assessment scores your shortlist on the same three axes. It usually adds one candidate nobody had listed, because it came from the operators rather than the strategy session.
Ready to scope AI Consulting?
Tell us where the work sits today and what is holding it up. We will come back with the shape of a first phase, what it would prove, and what running it takes.
+91 97915 97993