About
“The best system is the one your team never has to think about.” That line has guided every build since day one.
An AI and ERP company in Ahmedabad
Working with businesses across India, Canada and Australia.
2
Peer-reviewed papers
Published machine learning research behind the AI solutions we ship.
FDE
Forward-deployed engineering
No account managers, no handoffs, no juniors learning on your project.
NDA on request
Your data stays yours. Nothing you share is used for marketing.
Why we exist
Most enterprise AI never leaves the demo.
CodeCrafters started inside the systems most agencies avoid: ERPs with a decade of customisations, finance teams reconciling by hand, warehouses running on spreadsheets. We learned that the intelligent part of an AI project is the easy part. The hard part is the data underneath it, the platform around it, and the people who have to trust it on a Monday morning.
So we built a company that owns every layer. We design the agent and the pipeline, but we also fix the data model, harden the infrastructure, and sit with the operations team until the new workflow is boring. That is why our deployments are still running years later.
We stay small on purpose. Founder-led, senior-only, limited to 3 new enterprise engagements per quarter. It is the only way we know to keep the promise in the headline.
How we work
Four rules we do not bend.
Business first, code second
We start from the P&L line you want to move, then pick the technology. Never the other way round.
Compound value
Every system we ship is built to be extended. Month twelve should be cheaper than month one, not more expensive.
Boringly reliable
Exciting demos are easy. Quiet production is the job — systems still running years after handover, with no one thinking about them.
Forward-deployed, not thrown over the wall
The people on the diagnostic call are the people writing the code. No offshore relay, no juniors learning on your account.
Led by
Jaimin Shah
Founder & Principal Engineer · Ahmedabad, India
Research rigour, applied to production systems.
8+
AI systems in production
120+
Custom modules shipped
30+
Projects delivered
2
Peer-reviewed papers behind the work
CodeCrafters exists because most enterprise software is fragile, and the reason is almost never the software. It is the depth of thinking behind it. We were built out of years spent inside a major ERP vendor watching rollouts that technically shipped and never actually landed.
So we work the other way round. Every engagement is scoped and architected by the person who will build it, and that person stays on the account until it is running. No discovery team handing a deck to a delivery team. No handoffs.
The research background matters here in a practical way: peer-reviewed machine learning work means we know the difference between a model that demos well and one that holds up against real data. That is the same judgement we apply to an ERP rollout or an AI agent going near production.
Publications
Hybrid deep learning for stock market prediction on Nifty50
Elsevier · Intelligent Systems with Applications · 2022
doi:10.1016/j.iswa.2022.200111Noise-filtered EMG gesture recognition with CNNs
Taylor & Francis · 2022
Peer-Reviewed Publication
EMG Gesture Recognition · 2022
Co-authored a peer-reviewed paper on noise-filtered EMG signal processing for hand gesture recognition, reaching 95% classification accuracy with CNN architectures.
LLM Email Intelligence Pipeline
Zod-validated structured outputs · 2026
Independently architected a production Gmail intelligence system that classifies emails into structured life events using multi-provider LLM routing via Vercel AI Gateway.
OHIP Insurance Automation Platform
MCEDT Integration · Ontario, Canada
Built a real-time healthcare insurance platform with OHIP verification, automated XML/EDI claims processing and MCEDT integration, handling end-to-end medical billing workflows.
Server-Side AES-256-GCM Token Encryption
Replacing a browser-side E2EE model
Redesigned a SaaS auth model by migrating from browser-passphrase E2EE to server-side AES-256-GCM encryption, eliminating a class of client-side key exposure vulnerabilities.
Sales Forecasting Dashboard
+45% inventory planning accuracy · 2023
Built a Python + OWL time-series forecasting dashboard that cut inventory discrepancies by 60% and lifted planning accuracy by 45% across multi-warehouse operations.
Mentored 8 Junior Developers
50+ technical interviews conducted
Led onboarding, code reviews and growth paths for 8 junior developers and conducted 50+ technical interviews, building a scalable hiring and knowledge-transfer culture.
Measured
The numbers we can stand behind.
Projects delivered
Odoo modules shipped
ERP migrations delivered
Developers mentored
Peer-reviewed publications
Inventory planning accuracy
Stock discrepancies
Less manual data entry
Next step
Work with people who have shipped it before.
A free 45-minute diagnostic. We map your biggest bottleneck, estimate the ROI, and tell you honestly whether AI belongs there. No pitch deck.
Straight to the engineer. No form, no gatekeeper.
Accepting new engagements · Q4 2026