Healthcare / Insurance TPA · Delhi, India
MedAssist TPA Services
with Rajiv Nair, CTO, MedAssist TPA Services
01Problem
MedAssist was manually processing 1,800 insurance claims and 600 prior authorization requests per month across three insurer portals — each with a different UI, different data formats, and different submission rules. Average claim processing time was 22 minutes. A team of 11 was spending 80% of their time on data entry and portal navigation, with a 6.2% error rate triggering claim rejections and rework.
02Solution
- 01Built a hybrid Power Automate solution within MedAssist's existing Microsoft 365 tenant — unattended bots for standard claims, attended assist mode for complex or disputed cases
- 02Developed three separate bot workflows — one per insurer portal — each handling login, data extraction from the internal TPA system, field mapping, submission, and confirmation logging
- 03Deployed Terraform-managed Azure infrastructure with the bot runner in a private subnet, credentials in Azure Key Vault, and all bot actions logged to a centralized audit table
- 04Wired Azure Monitor alerts to PagerDuty with severity routing — P1 (bot down), P2 (queue breach), P3 (elevated exception rate)
- 05Integrated prior authorization workflow: bot submits requests automatically and polls for insurer response, escalating to human queue only on denial or additional-information requests
03Results
- 87% straight-through processing rate — 1,566 of 1,800 monthly claims handled end-to-end without human intervention
- Processing time reduced from 22 minutes to 3.5 minutes per claim — a 6.3x throughput improvement with the same headcount
- Annual savings of Rs. 38 lakh in processing labour — the team of 11 now handles 3x the claim volume with 4 staff redeployed to client relationship management
- Claim error rate dropped from 6.2% to 0.3% — rejection-driven rework eliminated
- Prior authorization turnaround: from 2–3 days to under 4 hours for standard requests
- Client
- MedAssist TPA Services
- Industry
- Healthcare / Insurance TPA
- Location
- Delhi, India
- Timeline
- Oct 2025 – Apr 2026
- Verified via
- Client testimonial
“Most RPA vendors treated infrastructure as someone else's problem. Techtweek handled the Azure setup, the bot deployment pipeline, and the actual automation — we didn't coordinate between two vendors. The prior auth bot alone saved us Rs. 18 lakh in the first year.”
related work
More engagements like this
Damsol was launching a crypto application for Indian users but lacked the in-house AWS expertise to host and run it reliably.
Scratchie needed a partner to take full ownership of its cloud servers and DevOps so the founding team could focus on the product.
CapiFin was processing 3,200 personal loan applications per month entirely manually — document collection, CIBIL pulls, eligibility calculation, and LOS data entry consuming 14 FTE-days per month. Month-end backlogs regularly hit 400+ applications, pushing approval TAT to 5–7 days and triggering borrower drop-off.